Western Forest Strategy Adaptive Management Summary

Forest Investment Account, Land Based Investment Program Resource Information Component – Monitoring Activity Area

Progress Report – 2009/10

W.J. Beese, MF, RPF Forest Ecologist

J.A. Deal, RPBio, RPF Strategic Planning Biologist

March 2010

Corporate Forestry 118 – 1334 Island Highway Campbell River, BC V9W 8C9 Acknowledgements

Our sincere thanks go to all of the managers, foresters, engineers, loggers and support personnel that take the ideas in the Western Forest Strategy and put them into practice. We thank the senior management of Western Forest Products for having the vision to carry on with this strategy through difficult economic conditions.

The authors are especially grateful to Glen Dunsworth, Fred Bunnell, Dave Huggard and Laurie Kremsater for laying the foundation for the Adaptive Management Program, and to Jeff Sandford for his important role in field coordination and database development. We also thank Nick Smith, Ken Zielke and Bryce Bancroft for their close collaboration on this effort, the many project leaders whose work is presented, and participants in various Working Groups over the years. We also thank Sue McDonald, Pat Bryant, Michael Fowler, Annette van Niejenhuis and Kathy Wood for their significant contribution “behind the scenes” to keeping many projects running smoothly.

Executive Summary

Western Forest Products is implementing a forest management strategy to sustain biological diversity within the company’s tenures. The Western Forest Strategy (WFS) has three broad goals:

1. Represent the full range of ecosystems within the non-harvestable landbase to maintain lesser known species and ecological functions. 2. Maintain structural attributes of older forests distributed across the landscape and in harvested areas to support biological richness. 3. Sustain productive populations of forest-dwelling species over time.

The WFS uses landscape zoning, variable retention and adaptive management to achieve these goals while balancing other ecological, social and economic objectives. Because the effectiveness of variable retention and broad landscape zoning in maintaining biodiversity is largely untested, “adaptive management” is a key component of the WFS. Adaptive management (AM) is a structured approach to learning from operational practices, monitoring and experiments to provide feedback to management and continual improvement.

This report summarizes progress on implementation of this strategy from 2005 through 2009, including significant findings from the monitoring and adaptive management program that supports the strategy. The WFS made significant progress toward achieving its goals:

1. Ecosystem Representation • We combined spatial data from the three legacy company tenures that are now covered by the WFS, including Terrestrial Ecosystem Mapping • We initiated a new analysis of representation across company tenures to be completed in the second phase of this project

2. Maintaining Old Growth Forests and Structural Attributes • We established OGMAs as part of landscape unit planning • We continued to implement the variable retention approach to forest harvesting, including us of the retention silvicultural system for the majority of the company’s harvesting over the past five years • We monitored forest structural attributes on experimental Variable Retention Adaptive Management (VRAM) and operational sites

3. Sustaining Species • We conducted monitoring and research projects on forest birds, amphibians, gastropods (slugs and snails), ectomycorrhizal fungi, carabid beetles, vascular plants and lichens • We continued development of a species accounting system for monitoring groups of species with different habitat needs

4. Adaptive Management (AM) and Monitoring • We revised the forest strategy, retention guidelines and zoning framework following changes in company ownership and tenure • We established a new Forest Strategy working group, developed retention guidelines and training materials, and conducted training sessions

ii • We monitored retention system implementation levels and established new standards for spatial identification of long-term retention • We continued with species and structure monitoring on operational cutblocks and nine VRAM experimental areas • We monitored the impacts of retention on forest growth and yield, windthrow and small streams

The forest strategy has helped the company achieve and maintain CSA Sustainable Forest Management Certification for several forest operations, supports marketing of WFP’s forest products and helps us maintain public support.

This report summarizes progress on implementation of the strategy and monitoring findings from the past decade. Part II will present results from an ecological representation analysis, describe how we have used the adaptive management process and present a 5-year plan for 2010 – 2014.

iii Table of Contents ACKNOWLEDGEMENTS ...... I

EXECUTIVE SUMMARY ...... II

BACKGROUND...... 1

RATIONALE FOR THE LANDSCAPE ZONING APPROACH ...... 3

OBJECTIVES...... 4 RATIONALE AND EXPECTED OUTCOMES...... 4 METHODS...... 5 STUDY AREAS...... 5 ANALYSIS OF ECOSYSTEM REPRESENTATION ...... 5 STAND-LEVEL RETENTION OF HABITAT ATTRIBUTES ...... 6 REVIEW AND SUMMARY OF SPECIES MONITORING AND RESEARCH...... 7 REVIEW AND SUMMARY OF SILVICULTURE MONITORING AND RESEARCH...... 8 DEVELOPMENT OF A REVISED AM PROGRAM AND 5-YEAR PLAN (PHASE II) ...... 8 DATA MANAGEMENT ...... 9 RESULTS...... 10 REVISED FOREST STRATEGY...... 10 IMPLEMENTATION MONITORING...... 11 ANALYSIS OF ECOSYSTEM REPRESENTATION ...... 15 STAND-LEVEL RETENTION OF HABITAT ATTRIBUTES ...... 15 REVIEW AND SUMMARY OF SPECIES MONITORING AND RESEARCH...... 18 Species Accounting System ...... 18 Overview of Broad Conclusions from Species Monitoring ...... 18 Vertebrate Species Monitoring ...... 19 Invertebrate Species Monitoring ...... 23 Vegetation Species Monitoring...... 25 Montane Alternative Silvicultural Systems (MASS) – vegetation studies...... 26 Other biodiversity studies at MASS include birds and canopy insects...... 27 Species at Risk Monitoring ...... 27 REVIEW AND SUMMARY OF SILVICULTURE MONITORING AND RESEARCH...... 29 Variable Retention Windthrow Monitoring ...... 30 Growth and Yield — Edge Regeneration Studies ...... 32 Montane Alternative Silvicultural Systems (MASS) – Regeneration Studies ...... 33 THE ADAPTIVE MANAGEMENT CYCLE...... 35

ADAPTIVE MANAGEMENT PLAN (2010-2014)...... 36

REFERENCES ...... 37

APPENDICES...... 40

Background

Western Forest Products is implementing a forest management strategy to sustain biological diversity within the company’s tenures. The Western Forest Strategy (WFS) has three broad goals:

The WFS uses landscape zoning, variable retention and adaptive management to achieve these goals while balancing other ecological, social and economic objectives (Bunnell and Dunsworth 2009). Because the effectiveness of variable retention and broad landscape zoning in maintaining biodiversity is largely untested, “adaptive management” is a key component of the WFS. Adaptive management (AM) is a structured approach to learning from operational practices, monitoring and experiments to provide feedback to management and continual improvement (Holling 1978, Stankey et al. 2005).

History of the Strategy

In recent years, changing public values and scientific knowledge have led forest practices towards achieving broader objectives on public forest lands in British Columbia (BC). Global concerns about biological diversity have raised demand in the marketplace for wood from forests under sustainable forest management certification. In 1998, MacMillan Bloedel (MB) announced “The Forest Project”—an innovative forest management strategy designed to achieve a balance of ecological, social and economic goals for managing mainly public forest land in coastal BC. The strategy included landscape zoning, increasing old-growth conservation, adopting variable retention harvesting, implementing a monitoring and adaptive management program, and achieving independent forest certification (Dunsworth and Beese 2000; Beese et al. 2001). In 1999, Canadian Forest Products (Canfor) adopted the “Forestry Principles” —an approach to sustainable forest management for their coastal region that included ecosystem management and nine other principles (Deal 2002). Weyerhaeuser bought MB in 1999 and continued the Forest Project as the “Coast Forest Strategy.” A decade later, after several changes in company ownership and mergers, these programs are continuing under Western Forest Products Inc. (WFP) as the “Western Forest Strategy” for managing over 1.4 million hectares of forest land on the BC coast.

Use of the variable retention (VR) approach to harvesting is a recent development in BC since the term was first introduced by the Clayoquot Scientific Panel (1995) and subsequently described in a broader context by Franklin et al. (1997). Prior to this time, there was very little experience in using partial cutting silvicultural systems in the old-growth forests of the BC coast (Arnott and Beese 1997). In 1999, the BC government officially recognized the “retention silvicultural system” in its forestry regulations. This new silvicultural system acknowledges that retention of trees and other structural attributes of forests for purposes other than tree regeneration and timber production are legitimate goals of forest management (Mitchell and Beese 2002). Weyerhaeuser purchased MB in 1999 and continued phase-in of the VR approach throughout its BC coastal operations over a 5-year period, completing over 90 percent

Western Forest Strategy – AM Summary 1

of harvesting using VR in 2003. Through 2006, virtually all harvesting was VR on the original public tenure of MB that became part of WFP. Field implementation of the Forestry Principles by Canfor began in 2001 and included specific targets for stand-level retention as patches or individual trees, depending on the landscape unit, biodiversity emphasis and biogeoclimatic variant. This summary describes how an adaptive management approach was used to apply the results of research and monitoring to a new landbase created through the merger of three companies in order to continue to achieve the principal goal: to sustain biodiversity, or biological richness and its associated values, within the company’s tenure.

Biological Rationale for Retention

Forest ecosystems and species have evolved in response to changes in climate and different natural disturbances at various scales. To achieve conservation of biological diversity, the basic theoretical premise is that species are adapted to historic local conditions. In coastal BC, windthrow, insects, disease, infrequent fire and landslides create forests with an abundance of dispersed residual structure (e.g., live and dead standing trees in varying patterns) from the pre- disturbance stand. Our approach is to use scientific knowledge of historical development and habitat as a guide to sustain productive and diverse forest ecosystems (Franklin et al. 1997, Beese et al. 2003). We recognize the resilience of ecosystems and the multiple pathways and patterns that can occur within the limits of ecosystem processes; therefore, we do not believe it is necessary to “mimic” natural disturbances. Our strategy assumes that both stand-level retention and landscape-level reserves are necessary for conserving biodiversity across the landscape (Lindenmayer and Franklin 2002). Neither approach alone is likely to be as effective or efficient. A comprehensive biological rationale for the strategy was described by Bunnell et al. (2003).

Coastal BC has a diversity of forest ecosystems and species; therefore, forest management practices must vary in response to that diversity. No single harvesting or silvicultural system is appropriate everywhere. Clearcut, seed tree, retention, shelterwood and selection systems are all ecologically appropriate in the right context. A mixture of systems will achieve a range of patch sizes and structures within stands and landscapes.

The Western Forest Strategy allows for a full range of silvicultural systems to be used for maintaining diverse forest conditions on the landscape. The approach to stand-level retention includes the following elements:

1. Maintain long-term retention on cutblocks using a variable retention approach; 2. Choose the most appropriate silvicultural system, based on safety, ecological and operational factors; 3. Implement guidelines and targets for the amount of the retention system used in each Resource Management Zone based on ecological units, site and stand conditions, economics and harvesting systems; 4. Encourage a diversity of approaches throughout the company but maintain consistent standards.

The retention silvicultural system (Mitchell and Beese 2002) is used for the majority of WFP’s overall harvesting applying the zoning strategy described above. We also use clearcut with reserves that typically retains as much area in long-term retention, but without the same spatial distribution. Although less visually aesthetic, clearcut with reserves is a means of retaining stand-level habitat where wind hazard or economics make the retention system difficult.

Western Forest Strategy – AM Summary 2

Rationale for the Landscape Zoning Approach

The basic concept behind zoning is to create different intensities of management in different landscapes under the assumption that a variety of conditions will support more effectively a variety of organisms. The other purpose is to recognize the ecological differences in the type and frequency of natural disturbances that have shaped the seral stage distribution and stand structural conditions in a landscape. Zoning is a useful framework for applying different regional or landscape-level management goals for the company’s tenure. The Western Forest Strategy uses existing BC Government ecological and administrative zoning as a framework to apply guidelines for the type and amount of stand-level retention across the WFP landbase. It combines management intensity zoning, ecological units and dominant disturbance types. The WFS zoning framework is a modification of the original approaches developed under the Canfor Forestry Principles (Deal 2002) and the MB – Weyerhaeuser strategy described in a recent book (Bunnell and Dunsworth 2009).

Adaptive Management and Monitoring

Monitoring is an essential component of Adaptive management (AM) and is composed of two main elements: performance and effectiveness1. Performance (or implementation) monitoring measures whether or not you have met the stated targets and standards. Effectiveness monitoring measures whether or not you have achieved the desired outcomes.

Key elements of the program include: • An AM framework with criteria and indicators to guide the program; • A Forest Strategy Working Group to provide input on implementation and monitoring; • Implementation monitoring on a random sample of cutblocks; • Establishing experimental areas for comparing the biological and economic impacts of variable retention. • Monitoring forest structural attributes for different approaches to variable retention and for benchmark sites; • Monitoring the impacts of VR on forest growth and yield, windthrow and small streams; • Monitoring forest birds at stand and landscape levels; • Completing pilot studies for a variety of species groups, developing indicator species for continued monitoring and designing a species accounting system; • Collaborated with government, universities, NGOs and other forest companies on research and monitoring initiatives.

Linking monitoring back to management action is a fundamental component of an effective operational AM program.

1 Some authors use the term “validation monitoring” to distinguish monitoring or research that tests cause-effect relationships. We use the term effectiveness monitoring broadly to include this type of monitoring.

Western Forest Strategy – AM Summary 3

Objectives

The objective of this report is to summarize progress on the goals and major indicators of Western Forest Products’ Western Forest Strategy and present a five-year adaptive management plan for monitoring, research and feedback to management practices. The specific objectives are to:

• Evaluate progress towards the three major goals in the strategy for biodiversity conservation: representing the full range of ecosystems in unharvested areas, maintaining stand-level habitat attributes and sustaining forest-dwelling species; • Report the results of an analysis of ecosystem representation in the non-harvestable landbase on WFP TFLs (6, 19, 25, 39, 44); • Summarize stand-level retention of habitat attributes based on stand-structure monitoring and the proportion of different approaches across the tenure; • Summarize monitoring and research study findings for various species and species groups from projects conducted under the Adaptive Management program of the Western Forest Strategy since 1998 (including activities under the MB Forest Project, Canfor Forestry Principles and Weyerhaeuser Coast Forest Strategy), and other pertinent literature as time permits; • Summarize key findings from monitoring forest regeneration and windthrow and the implications for sustainable forest management; • Prepare a new Adaptive Management Plan to identify monitoring, research, extension and training needs for the next five years;

This document (Part 1) is a summary of progress and key findings. A separate document (Part 2) to be prepared by June 2010 will present the ecosystem representation analysis and five- year Adaptive Management Plan.

Rationale and Expected Outcomes

Adaptive management is a process of continual improvement. This report helps close the “feedback to management” loop in the AM cycle in the following ways: 1. We evaluate whether or not current strategies and practices are achieving biodiversity conservation and forestry objectives, as measured using indicators defined in the Western Forest Strategy. 2. We communicate the findings of several monitoring activities, including forest structure, windthrow, regeneration edge effects, birds and other species groups and summarize forest management impacts. 3. We report on a “coarse filter” biodiversity assessment utilizing inventory data from Terrestrial Ecosystem Mapping and identify areas in need of improvement. 4. We summarize species research funded by the Forest Science Program as a “fine filter” biodiversity assessment and suggest future monitoring strategies. 5. We will use the information summarized in this report to inform revisions to strategies, indicators, standards and practices.

WFP’s adaptive management program goes beyond the status-quo requirements of both the BC Forest and Range Practices Act and Canadian Standards Association (CSA) Sustainable Forest Management (SFM) Certification. Our program is one of the few examples of successful implementation of an adaptive management approach in forest resource management

Western Forest Strategy – AM Summary 4

(Marmorek et al. 2006). The outcome of this program is to further increase our knowledge of how to manage coastal forests with varying amounts and patterns of retention to meet silvicultural and biodiversity conservation objectives. We expect to produce results and recommendations that can be used by operational foresters for decision-making. The expected outcome “bottom line” is a science-based approach to sustainable forest management.

Methods

Study Areas

The WFS Adaptive Management program applies to Western Forest Products’ TFLs 6, 19, 25, 39 and 44 on Vancouver Island and the mainland coast (see map in Appendix 1). For purposes of this evaluation and plan, areas on Haida Gwaii (QCI) and the Central Coast that are managed under Ecosystem Based Management (EBM) land use plans, government orders and guidelines are excluded. The limited area of WFP private lands are also excluded from the analysis, although some of the monitoring and research over the past decade was conducted in part on private lands where results are applicable to similar Crown lands.

The extensive or “passive” AM framework consists of monitoring structure and the presence or absence of species along with forest growth, health and windthrow in current and future cutblocks. The intensive or “active” AM framework consists of five designed comparisons replicated three times and focused on specific stand-level questions using an experimental approach. We call these Variable Retention Adaptive Management (VRAM) sites; details of the design are described in Appendix 2. Treatments are operational scale (20-hectare units), so each area covers 80 to 100 hectares. Nine of fifteen sites have been established to date.

Analysis of Ecosystem Representation

Our broad “coarse filter” indicator for monitoring changes to the forest landscape that affect biological diversity is the proportion of each ecosystem that will remain unharvested. Many species, especially those for which we have little or no information, are best accommodated by ensuring that some portion of each distinct habitat type or ecosystem is represented in a relatively unmanaged state. Most unmanaged area on the BC coast is old growth forest (>250 years old). Unmanaged areas also provide an ecological baseline against which the effects of human activities can be compared—a critical role in the long-term monitoring required for adaptive management in forestry. The important question for monitoring is whether most or all ecosystem types are represented in the unmanaged land base and, secondarily, whether the size, shape, age and spatial distributions of unmanaged areas are appropriate.

WFP technical staff and contractors assembled a GIS spatial dataset for the company’s tenures including the forest cover, Resource Management Zones, Ecosections, biogeoclimatic units, Terrestrial Ecosystem Mapping (TEM) and the non-harvestable landbase (e.g., various reserves such as Old Growth Management Areas, Ungulate Winter Ranges, Wildlife Habitat Areas, Riparian Reserve Zones, Wildlife Tree Retention Areas, etc.). This fully intersected GIS coverage was used to generate reports of non-harvestable forest cover by ecosystem, using the hierarchy of ecosections, biogeoclimatic units (zones, subzones and variants), and site series to indicate the contribution of these areas towards ecological representation. A summary of the data needs and general procedures for the analyses are given in Appendix 3. Delays in

Western Forest Strategy – AM Summary 5

availability of GIS data postponed the final report on the analyses until Phase II of this project. The final report will describe areas of deficiency in ecological representation to help inform forest management planning and strategies for addressing these gaps in the future.

Stand-Level Retention of Habitat Attributes

Our “medium filter” approach to monitoring changes to the forest landscape affecting biodiversity is the amount and characteristics of stand structural elements that are retained after forest harvesting. Species whose habitat requirements are known can be accommodated by ensuring that these elements are well distributed in the managed forest landscape. We have five broad groups of structural variables in the monitoring program: 1) a standard set of habitat elements and their attributes, 2) more integrative habitat variables describing habitat structure, 3) processes critical to making long-term habitat projections, 4) broad-scale summary variables, and 5) indicator or idealized organisms. The standard set of structures being monitored include:

• Live trees – species, diameter, height, height of live crown (used for vertical structure), trunk breaks, and visible disease or pathogens. • Snags (standing dead trees) – species, diameter, height, decay class, and top breakage • Coarse woody debris (CWD) – species, size, decay class • Canopy – cover, depth and composition. • Shrubs – cover and composition • Ground cover layers – cover, height and composition of forb, bryophyte/lichen, litter/duff and inorganic layers

Other habitat variables are derived by combining several measured elements in a plot or a stand, while some require additional sampling or different survey methods (Huggard 2006). We have developed computer models that use data from structure monitoring (including growth and decay information) to project habitat supply for selected species over time.

We examined the results of stand structure monitoring, including post-harvest changes after 5 years, to evaluate how effective current practices and guidelines are at achieving distribution of habitat elements across our tenures. Areas with negative trends or that need improvement were described to help inform forest management planning and practices for addressing these issues. Analyses of data from 1999 through 2004 were summarized in Bunnell and Dunsworth (2009). Work completed since this synthesis is summarized in Table 1. The latest complete update for habitat structure monitoring was done in 2006 (Huggard 2006). A comparison of group and mixed retention in operational cutblocks across the three legacy companies was done the following year (Huggard 2007). Ongoing measurement of experimental VRAM sites continued in 2008 and 2009, with final 5-year post-harvest measurements planned for two VRAM sites in 2010. An analysis of all nine VRAM sites will be completed in 2010.

The six major questions to be addressed by habitat structure monitoring are:

1. How well do different types of retention (e.g., group, dispersed) retain different habitat elements? 2. How do retention patches compare to uncut “benchmark” stands? 3. How does the percent retention affect different habitat elements? 4. Are there significant edge effects? 5. How do retained elements vary by patch location (e.g., wetlands, rock outcrops, typical mesic sites)? 6. Has retention of different elements changed over time in operational cutblocks?

Western Forest Strategy – AM Summary 6

Table 1. Summary of habitat structure monitoring

Year Monitoring and analysis 2004 Re-measurement (5-year post-harvest) of 1999 sites and additional benchmark sites; updated summary report for 1999-2004 (193 VR blocks, 98 benchmarks, 52 other sites), edge effects analysis 2005 Re-measurement (5-year) of 2000 sites; updated summary report including analysis of 72 sites with 5-year post-harvest data, additional edge effects analysis 2006 Re-measurement (5-year) of 23 operational blocks from 2001 and 8 benchmark sites; measurements at 3 VRAM sites; updated summary report 2007 Measurement of 22 operational blocks from CWHxm and CWHvm subzones to compare group and mixed retention by 3 legacy companies, and compare to xm (30) and vm (37) benchmark sites; comparison of 1999, 2000-2003 and 2007 periods 2008 Re-measurement (5-year) of one VRAM site 2009 Re-measurement (5-year) of 4 VRAM sites Two more VRAM sites to be measured in 2010, plus analysis of all 9 VRAM sites 5-years post-harvest

For this report, we extrapolated findings from monitoring to the WFS landbase using annual reports of the area harvested by silvicultural system, implementation monitoring (Bancroft and Zielke 2004), windthrow monitoring (Rollerson et al. 2008) and the revised zoning standards.

Review and Summary of Species Monitoring and Research

Our “fine filter” approach to monitoring changes to the forest landscape affecting biodiversity examined selected forest-dwelling organisms. Success in sustaining species is the ultimate measure of success of the commitment to sustain biological diversity; however, it is impossible to monitor all species, so choices must be made thoughtfully. Three broad features guided the choice of species: sensitivity to forest practices, ease of monitoring, and utility of information to guide management (Kremsater et al. 2003).

Through pilot studies we attempted to assess how costly it would be to acquire information with the requisite precision and accuracy, and weighed that against the relevance of the information to adaptive management. Our AM framework document detailed specific organisms to monitor (Bunnell et al. 2003). The list of potential species includes: vascular plants (specific groups for canopy cover, downed wood, old forests or forest interior, rare species, exotics, streamside and those with specific forest hosts), bryophytes (ground and lower stem), lichens (canopy epiphytic), macro-fungi (ectomycorrhizal), terrestrial vertebrates (birds, shrews, amphibians), and invertebrates (beetles, slugs and snails). We have assumed that microscopic organisms are being monitoring indirectly by monitoring other aspects of ecosystems (e.g., tree and plant growth, organisms higher in the food chains).

Species monitoring projects were summarized in Beese et al. (2005a) and preliminary results for several species groups were presented in Bunnell and Dunsworth (2009). Table 2 lists the species groups that have been studied over the past decade as part of the company’s research and AM program (not including squirrels; dropped after pilot studies). Most of these projects were funded by the Forest Investment Account (FIA) Forest Science Program or FIA Land

Western Forest Strategy – AM Summary 7

Table 2. Summary of species monitoring and research

Topic Recent reports and papers Animals Birds – landscape level Preston and Campbell 2009 Birds – stand level Preston and Harestad 2007, Chan-McLeod 2008, Manning 2008 Amphibians Wind 2008 Carabid beetles Pearsall 2008 Gastropods Ovaska and Sopuck 2008 Stream invertebrates Paulson 2007 Plants Ectomycorrhizal fungi Outerbridge and Trofymow 2004, 2008 Lichens and bryophytes Stanger 2004, Sadler 2004 Vascular plants Beese et al. 2009 (MASS)

Based Investment Program (LBIP), followed recognized scientific methods and applicable standards and were submitted to the applicable BC government library or data warehouse.

We examined the results of these studies and assembled the key findings relevant to forest management. When possible, we present species response curves to indicator variables (e.g., percent retention), trends over time or comparisons to some benchmark condition. Within the time constraints of the project, we also included relevant findings from a few other studies in coastal BC or neighbouring geographical areas, particularly from literature reviews or syntheses (e.g., Aubrey et al. 2004, Clayoquot Scientific Panel 1995, Coast Information Team 2004, Huggard and Bunnell 2007, McLennan and Hennon 2005, Rosenvald and Lõhmus 2008).

Review and Summary of Silviculture Monitoring and Research

Another important aspect of the AM program besides biodiversity was to monitor the impacts of management practices on forest regeneration, growth and health. Projects undertaken include: • Windthrow monitoring (Rollerson et al. 2008); • Regeneration edge and retention effects (Smith 2009a,b, Iles and Smith 2006); • Montane Alternative Silvicultural Systems (MASS) studies (Mitchell et al. 2004, 2007; Beese et al. 2009).

We examined the results of these studies and assembled the key findings in our summary. Results will be used to assess trade-offs among competing resource management objectives.

Development of a revised AM Program and 5-Year Plan (Phase II)

We will convene a small workshop with WFP staff, colleagues from government agencies (MFR, MOE, ILMB) and consultants who have guided the strategy and AM program since its inception (Bunnell, Dunsworth, Huggard, Kremsater) to: • Review this document for errors or omissions; • Discuss synthesis, modelling, species accounting system development; • Brainstorm ideas for the AM 5-year plan; • Discuss partnership and funding opportunities.

Western Forest Strategy – AM Summary 8

Results of the workshop will be used to revise this report (Part 1) and prepare an updated AM program and five-year plan (Part 2). A summary of the workshop will be included in Part 2.

We will use the analyses and summaries of ecosystem representation, stand-level retention of habitat attributes, species and silviculture studies to prepare a revised adaptive management program and five-year plan. We will summarize the changes to the original Forest Strategy and role of the AM program over the past decade. We will assess the role of active experimental comparisons (Beese et al. 2005b) and passive operational monitoring in the AM program and develop a research and monitoring plan. Although the monitoring plan will not include individual project-level details (e.g., sample design), we will use the FIA standards for planning and reporting to guide the five-year plan document. We will also use the “Checklist for Biodiversity Monitoring Projects” as the framework for the plan. We will describe the species accounting system under development and our approach to modelling and predicting future trends. We will list the key issues that may require adjustments to practices and strategies and discuss options. We will also identify information needs for operational planners (engineering, forestry) and forest workers. This project will allow us to complete the final step in the adaptive management and monitoring cycle: feedback to management.

Data management

All project data for WFP’s research and monitoring program is managed using an MS-AccessTM database. Data include: descriptive metadata, spreadsheets, reports, statistical analysis output, GPS locations, ArcMapTM spatial data and attributes. We also maintain digital images and physical samples (e.g., species collections). Digital data are stored in multiple locations with regular back-up (i.e., WFP intranet servers, external hard drives, CDs or DVDs). We also maintain a hard-copy backup of reports in WFP archival storage.

In addition, most of the projects included in the AM program were funded through the Forest Investment Account (Forest Science Program or Land Based Investment Program), were tracked in FIRS and RIMS, and project deliverables were submitted to the applicable BC government library or data warehouse. Individual project leaders (mostly consulting biologists) also maintain personal records of data, reports and publications, and specimens.

Western Forest Strategy – AM Summary 9

Results

Revised Forest Strategy

As a result of company mergers, Western Forest Products Inc. (WFP) now manages nearly 1.5 million hectares of productive forest land on the BC coast. Most of this area is public (Crown) land managed under Tree Farm Licenses (TFL) granted by the BC government. The three former companies—WFP, Canfor and Cascadia (formerly Weyerhaeuser)—had different approaches to forest management, particularly in the application of variable retention harvesting. Differences arose from several factors including the size of operations (e.g., Canfor’s TFL37 was less than 30% of the size of the other legacy companies), their location (e.g., WFP’s TFL6, TFL19 and Forest Licences on the windward, northwest portion of Vancouver Island) as well as the economic circumstances and overall management philosophy of the different companies.

The Western Forest Strategy (WFS) brought operations from the legacy companies together under a common strategy that supports WFP’s commitment to Sustainable Forest Management (SFM), including the goal to conserve biodiversity on managed forest lands. The WFS was approved by senior management in July 2007 (Appendix 1).The strategy builds upon the experience and success of the legacy companies while recognizing the value of a diversity of approaches in different geographical areas and forest types. The WFS landbase is significantly different from the MacMillan Bloedel landbase for which the original “Forest Project” strategy was developed. The major shift in the landbase was removal of private land tenure on the southeast portion of Vancouver Island (now Island Timberlands) representing dry biogeoclimatic units and an increase in the proportion of wet, windward biogeoclimatic units on northwest Vancouver Island.

Phase II of the AM Summary and 5-Year Plan will describe the adaptive management process and the changes to the Forest Strategy zoning, use of the retention system and guidelines. A brief description is given here to provide some context for the summary of monitoring results that follows.

The Western Forest Strategy uses existing BC Government ecological and administrative zoning as a framework to apply guidelines for the type and amount of stand-level retention across the WFP landbase (Table 3). It combines management intensity zoning under the Vancouver Island Land Use Plan (i.e., Enhanced, General, Special), ecological units and dominant disturbance types (represented by Ecosections, and biogeoclimatic subzones and variants). This approach is similar to the zoning applied by Canfor in TFL37 (Deal 2002), which developed retention targets and guidelines based on landscape units, biodiversity emphasis options, biogeoclimatic subzones and disturbance history. Table 3 shows the approximate WFS equivalents to the Coast Forest Strategy stewardship zones implemented on former Weyerhaeuser BC Coastal tenures.

The Resource Management Zones of the Vancouver Island Land Use Plan (VILUP 2000) were used as the next level of stratification within Ecosections for guiding stand-level retention

Western Forest Strategy – AM Summary 10

Table 3. Comparison of original and revised zoning approach

% Long- Original CFS % Retention Term Zone New WFS Zone System Retention Enhanced Windy 30 10 Timber Enhanced 50 10 Enhanced Dry 60 15 General Windy 40 15 Habitat General 60 15 General Dry 70 20 Old Growth Special (see VILUP) 20 CFS=Coast Forest Strategy; WFS=Western Forest Strategy % Retention system: percentage of logged hectares in Zone using the retention system. % Long-term retention: minimum long-term retention within a retention system cutblock. All original zones required 100% variable retention (retention system or non-clearcut) Long-term retention for the CFS zones: Timber 10%, Habitat 15%, Old Growth 20% Original Old Growth zone required multi-aged systems, with 2/3 of zone in reserves

practices. Enhanced Management Zones (EMZ) have the lowest retention requirements and highest timber expectations. General Management Zones (GMZ) use intermediate levels of retention and greater use of the retention silvicultural system. Special Management Zones (SMZ) are mandated by government to use the retention system or clearcuts up to 5 hectares.

Within the EMZ and GMZ, guidelines vary by ecological unit with lower targets for the retention system in windthrow prone areas. The strategy results in considerable variation in the use of the retention system (from 30% to 100% across strata) and variation in the minimum stand-level retention (10% to 20%) across WFP tenure.

Implementation Monitoring

Performance monitoring is key to achieving the Western Forest Strategy objectives. It allows us to answer the questions: “Did we do what we said we would do? Are we meeting our own goals and guidelines?” Through annual spreadsheet reports submitted by each WFP forest operation, we track the percentage of harvest area under the retention system (and other silvicultural systems) and the amount of retention left to meet the standard for the zone.

Through independent audits (Bancroft and Zielke 2004), we assessed whether or not individual cutblocks were meeting the goals for variable retention during the initial 5-year phase-in of this strategy under MB and Weyerhaeuser. Results of performance monitoring representing a random sample of 17% of the harvested area (248 cutblocks; 7,150ha TAUP; 4797 ha NAR)2 showed that on this portion of the current WFP landbase: • about two-thirds of blocks were rated as good to excellent examples of VR; • assessments showed improvement in both prescriptions and implementation (key areas included visual design, marking danger trees and avoiding leave-tree damage);

2 TAUP=Total Area Under Prescription; NAR=Net Area Reforested (i.e., cut area).

Western Forest Strategy – AM Summary 11

• retention was judged to be of good to optimal choice to provide a range of wildlife habitat, such as riparian features, deciduous trees, dens and nests, snags.

Symmetree assessed the type of attributes retained in 1,591 groups. They found a wide range of group sizes were used with an average of about 1 ha. Snags occurred in 78 to 80% of the groups in 2002 and 2003. When present, groups were “anchored” on special features such as bear dens, nest trees, culturally modified trees and large veteran trees. Riparian features (28- 43%), rock outcrops (14-16%) and deciduous trees (7-16%) were also used frequently as anchors for retention patches (2001-2003).

Retention along small streams that do not require treed buffers under BC regulations was assessed along 151 km of stream length within VR cutblocks in 2001 through 2003. Some retention occurred along 76km (50%) of these streams and 51 km (34%) had retained buffers 20 m wide or greater along the stream edge. An additional 42 km of streams were along block boundaries and 30 km (71%) had at least 20 m of retention buffering the stream.

Because of successful results, independent audits of a random selection of cutblocks were discontinued after 2003. The plan was to conduct audits in 2009 from a sample of cutblocks completed from 2004 through 2008. In 2004, assessments were done to monitor a selection of cutblocks completed under ‘large patch VR’ and ‘standing stem harvesting’ guidelines (Bancroft and Zielke 2005a,b). Due to financial constraints, no independent performance monitoring has been done in recent years.3

We also tracked overall performance during the initial 5-year phase in of the strategy. Results showed that legacy Weyerhaeuser areas: • achieved over 90% variable retention in 2003; • completed 27,671 ha of VR over five years (exceeding the goal); • improved Timberlands safety performance (industry leading) during VR phase-in; • exceeded the minimum long-term retention requirements for each cutblock; • met or exceeded phase-in targets every year except the final year.

Company operations continue to complete an annual spreadsheet report summarizing all harvested cutblocks by silvicultural system for tracking implementation of the strategy. Figure 1 summarizes use of variable retention over the past decade on the legacy Weyerhaeuser portion of the tenure, primarily TFL39 and TFL44 (data through 2004 include private lands now owned by Island Timberlands). Implementation of VR continued unchanged in 2005 after the Crown portion of former Weyerhaeuser operations became Cascadia Forest Products, and after the purchase of Cascadia by WFP in 2006. Therefore, virtually all harvesting in legacy Weyerhaeuser operations was done using the retention system during 2005 and 2006. With approval of the new strategy in 2007, use of the retention system in TFL’s 39 and 44 began a phase-in towards the estimated amount for these areas (57%). This is an estimate because the targets are applied by zone (Table 3), not by management unit (see Appendix 1). The phase-in goal for 2009 was to be half-way towards the zone targets (legacy WFP operations trending up, Weyerhaeuser trending down, and Canfor remaining about the same), with 2008 at the discretion of each operation. Including all WFP areas under the WFS, the percentage of area harvested using variable retention in 2008 was 48% (and was estimated at 55% for 2009).

The long-term retention level in each cutblock has averaged over 20% over the past decade (Figure 2), which includes group and dispersed retention and other reserves (e.g., riparian,

3 Not including audits done for CSA Z809 SFM Certification, ISO14001 EMS, the BC Forest and Range Evaluation Program, or Forest Practices Board.

Western Forest Strategy – AM Summary 12

100

90 % VR Target 80

30 % of Harvested Area

0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Figure 1. Variable retention over a 10-year period on the legacy Weyerhaeuser portion of the tenure, primarily TFL39 and TFL44. VR includes the retention system, and shelterwood and selection systems with reserves.

wildlife tree patches) within the cutblock boundaries. This meets or exceeds the 10% to 20% minimums required for all WFS zones (Table 3). A range of retention levels, including both short- and long-term retention were used in cutblocks. Most cutblocks had retention in the 11 to 30% range. Our practices show an intentional concentration of retention at the lower end of the range that is designed to maintain structural attributes while attempting to minimize the predicted reduction in growth and yield. We have a sufficient number of cutblocks throughout a wider range (5% to 50%) to assess impacts in our monitoring program.

Most of the variable retention harvesting was done as group retention or mixed (a combination of groups with some dispersed trees, Figure 3). Few cutblocks were done exclusively as dispersed retention. The proportion of dispersed blocks decreased from 21% in 1999 to under 2% in 2003, and very little in the last 5 years. This reflects a number of factors, including the shift in landbase away from second growth forests on SE VI where dispersed retention is more feasible as well as fewer dispersed prescriptions because of the generally higher incidence of windthrow compared to group retention. Over the entire decade, about 22% of cutblocks used a mixed approach; however, the proportion of mixed blocks was only 11% over the past 5 years. Shelterwood and selection systems with reserves were used for a minor portion of our harvest (<2%). A very small amount of harvesting was done using the “single stem” technique, whereby a helicopter removes a cut and limbed tree without it falling to the ground. This technique has applications on very sensitive terrain, or as a first pass removal of high value stems prior to conventional yarding.

Western Forest Strategy – AM Summary 13

10 % Long-term Retention

0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Figure 2. Long-term retention levels for all VR blocks from 1999 – 2008, based on annual reports from all forest operations.

12000 Other Dispersed 10000 Mixed Group 8000

6000 Hectares 4000

2000

0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Figure 3. Area harvested under different variable retention approaches over 10 years. “Other” includes patch cut, shelterwood or selection systems (with long-term reserves).

Western Forest Strategy – AM Summary 14

Multi-pass harvesting was used on roughly 10% of all cutblocks from 2002 through 2005, but much less in recent years. The most common usage of two or more harvest entries was where a cutblock had adjacency restrictions under the Forest Practices Code that required retention of at least 40% of the basal area of the stand—a rule designed to prevent progressive clearcutting. In other cases, windthrow or visual concerns were the rationale for a two-pass approach. Under the Forest and Range Practices Act (FRPA), applying the retention system using specific standards for spatial distribution is sufficient to avoid the 40 ha maximum cutblock size and restrictions on the timing of adjacent harvests.

Analysis of Ecosystem Representation

In Bunnell and Dunsworth (2009), results of an ecosystem representation analysis at the biogeoclimatic variant level for the legacy Weyerhaeuser portion of WFP are presented in Chapter 7 “Learning from Ecosystem Representation” (Huggard, Kremsater and Dunsworth, pp. 100-116). Briefly, this work concluded that overall, about 26% of the forested area was fully constrained from harvesting, while an additional 7% was partially constrained. The low elevation, dry biogeoclimatic units (e.g., CWHxm1/2, CWHdm) had consistently low representation of forests in an unmanaged state. As well, high productivity sites had, in general, less representation in the non-harvestable landbase than cooler, wetter, lower productivity ecosystems. The analysis also examined the original “Old Growth” zones and found that they contributed only small amounts to the representation of ecosystems that are most extensively managed. Old Growth zones were located in areas where there was a relatively large amount of remaining old growth, not in areas with dry biogeoclimatic units where old growth is rare.

The change in landbase resulting from the merger of three coastal forest companies (Weyerhaeuser (Cascadia), Canfor and legacy Western Forest Products) necessitated a new analysis of ecosystem representation. We repeated the analysis at the biogeoclimatic variant level and also added a new level of detail using site series (plant associations). WFP is fortunate to have nearly complete coverage of Terrestrial Ecosystem Mapping (TEM) for its tenures. This gives us the relatively unique ability to do representation analysis at the ecosystem level compared to other areas in BC.

As described above under Methods, we completed assembly of GIS data and initial summaries of the proportion of ecosystems in the harvestable and non-harvestable landbase; however, the final analysis and report on this work was delayed until the second phase of this project. We expect the analysis to show similar issues with representation of dry biogeoclimatic units and high productivity sites.

Stand-Level Retention of Habitat Attributes

In Bunnell and Dunsworth (2009), results of an analysis of monitoring stand structure are presented in Chapter 9 “Learning from Habitat Elements” (Huggard, Sandford and Kremsater, pp. 152-172). Briefly, this work reached the following conclusions for the original six questions posed [with notes added where subsequent work affects their conclusions]:

1. How well do different types of retention (e.g., group, dispersed) retain different habitat elements? Group retention had higher levels of retention of many habitat elements compared to dispersed retention, though dispersed did better at retaining large diameter live trees,

Western Forest Strategy – AM Summary 15

large but short snags and a few other variables. Mixed retention was generally similar to group retention though variable among different forest types. These findings highlighted the need to maintain both group and dispersed retention as useful VR tools.

2. How do retention patches compare to uncut “benchmark” stands? Large trees, overall basal area and some vegetation layers were retained at lower levels in retention patches compared to benchmark sites, while deciduous elements were more abundant in retention. Differences likely reflected more low productivity sites and wetlands in retention patches.

3. How does the percent retention affect different habitat elements? The abundance of habitat elements was proportional to percent retention for many tree and snag elements in group retention but more variable in dispersed retention. Amounts of coarse woody debris (CWD) and most vegetation layers were generally independent of percent retention in both systems.

4. Are there significant edge effects? Most habitat elements showed no effect of distance from patch edge into retention patches or into the harvested opening in surveys shortly after harvest. A few variables showed very minor decreases near the edge of patches. Only tall snags had enough of a decline near the edge to have any implication for the size and shape of retention patches. [Windthrow monitoring of cutblocks after several storm seasons, however, showed significant damage to retention patches with a higher percentage of damaged trees in small patches and strips compared with larger patches (see below, Rollerson et al. 2008)].

5. How do retained elements vary by patch location (e.g., wetlands, rock outcrops, and typical mesic sites)? Retention patches anchored on wetlands had the same or elevated levels of most habitat elements compared to typical forest, while patches anchored on rock outcrops had substantially reduced levels of many elements. Patches anchored on specific wildlife features differed little from other forest patches. These findings suggest that while variety in anchor types is useful, retention patches on wetlands can maintain greater amounts of habitat elements compared to other anchors.

6. Has retention of different elements changed over time in operational cutblocks? There were increases in some habitat elements during the first few years of implementation compared to the first year; however, most variables returned to initial levels in subsequent years. Annual variability in the characteristics and location of cutblocks sampled made it difficult to draw meaningful conclusions about progress. [Performance monitoring documented a significant trend of increasing average patch size retained over the first 5 years of implementation (Bancroft and Zielke 2004). Recent structure monitoring comparisons suggest that there are differences among forest operations in the type and amount of different habitat elements retained.]

The 5-year post-harvest summary for all nine of the VRAM sites will shed additional light on these questions. Unfortunately, this analysis will not be done until the end of 2010.

A recent analysis of a selection of cutblocks in the CWHxm and CWHvm biogeoclimatic variants compared 50 habitat elements in the operational blocks to levels in benchmark sites in uncut forest. The results provide a baseline for looking at trends in the future, and suggest several weak points in habitat retention that could be a focus for operational improvement:

Western Forest Strategy – AM Summary 16

• Retaining higher densities of large trees in both the CWHxm and CWHvm; • Retaining some very large snags, even if these are relatively well-decayed and short veterans from a previous stand; • Maintaining mid-size to moderately large snags in CWHxm; • In CWHvm, where snag retention was higher, emphasizing important rare features, like big hollow snags; • Continuing efforts to retain tall snags in CWHxm, which have declined over time; • CWD was generally retained or created at high levels, except for soft wood in CWHvm in general, and in CWHxm on sites with considerable ground disturbance during harvesting.

Many of these specific habitat elements are associated with higher productivity mesic sites, and may be missed if there is too much reliance on anchoring patches on wetlands or rock outcrops. Ensuring that some retention patches are located in productive mesic sites should be a priority.

Additional details on specific habitat elements are given by Huggard (2006, 2007).

Fungal inoculation of trees for habitat enhancement in second-growth forests

In addition to monitoring habitat structure in variable retention cutblocks, work was also initiated to investigate options for addressing potential habitat deficiencies in future forests. In some areas, especially where forest harvesting has been practiced for many years and second-growth stands are predominant, there is often a shortage of suitable wildlife tree habitat. Retention patches typically require removal of existing snags for safety of forest workers. Healthy, second- growth stands containing trees from 50 to 100 years of age can still take many decades before developing the primary wildlife tree attribute of heart rot. However, this natural process can be significantly accelerated through fungal inoculation. In these situations, artificial creation methods may be warranted in order to recruit wildlife trees more quickly than would otherwise occur through natural cycles.

In 2002, an operational trial to create wildlife trees in 60-year-old Douglas-fir stands using fungal inoculation was initiated in TFL 37 and TFL 44. In 2007, 50 trees were sampled to evaluate the effectiveness of two methods of inoculation: manual drilling and shooting an infected plug into the tree using a rifle.

Both the climbing/drilling and rifle inoculation methods were successful in introducing heart rot decay into live host trees and can be effective wildlife tree creation and enhancement tools (Manning 2007). The rifle method will take longer to produce useable decay columns, but this technique is much faster and therefore cheaper to conduct (i.e., depending on site factors, the rifle method is approximately 6 times faster and 47% cheaper than the climbing/drilling method.

Western Forest Strategy – AM Summary 17

Review and Summary of Species Monitoring and Research

Species Accounting System

WFP’s AM program includes species monitoring to determine whether existing forestry planning and practices appear to be successful in sustaining biodiversity. It is not feasible or affordable to monitor all species; therefore, an approach is needed that will help focus monitoring on species of concern that are likely to be affected by forest practices. Two approaches to help evaluate the effectiveness of biodiversity conservation strategies have recently been developed by Dr. Fred Bunnell and colleagues. One is a Species Accounting System that assigns species to monitoring groups. These groups combine natural history and features of monitoring to assign species to the most cost effective monitoring approach. Groups range from habitat generalists that require no monitoring to indicator species with specific habitat needs that are important candidates for monitoring. Species linked to habitat conditions that are easily monitored using GIS data are also grouped. The second approach includes an analysis of the impact of coarse- filter practices on key habitat attributes, including cavity sites, down wood, hardwoods, and shrubs. Practices that fall under the umbrella of coarse filter practices include general habitat management approaches, such as landscape level reserves, riparian reserves, stand retention, brush control, etc. Combining these two new approaches will yield both a robust, cost-effective monitoring approach and an effective way of focusing adaptive management on key questions.

A report was completed in March 2009 that tested the species accounting approach for TFL6 on northern Vancouver Island. The objectives of this project were to modify the species accounting system developed elsewhere in BC to accommodate coastal forest types and variable retention. Species likely occurring within TFL6 were assigned to modified monitoring groups with the idea that this system would expand to other WFP tenures in subsequent years if successful. Tests were largely limited to bird data--the richest component of vertebrate diversity. The study expanded the species accounting system to accommodate non-vertebrates to the extent possible. The study identified species for which habitats are too highly specialized or localized to accommodate coarse-filter approaches. The project also combined the modified species accounting system with recently developed analyses of coarse filter approaches to complete a sustainability analysis for late seral, cavity sites and down wood.

The following sections report findings from species monitoring over the past decade that have helped in development of the species accounting system and suggest species and approaches for future monitoring.

Overview of Broad Conclusions from Species Monitoring

In Bunnell and Dunsworth (2009), initial findings from monitoring different species are presented in Chapter 11 “Learning from Organisms” (Huggard and Kremsater, pp. 219-237). The authors note that monitoring organisms proved to be the most challenging of the three broad indicators of biodiversity (i.e., compared to ecosystem representation and forest habitat structures). Some general conclusions for three important questions are summarized below.

1. What types of VR are most useful for maintaining organisms? Retention of patches appears to be sustaining some organisms commonly assumed to be dependent on later seral stages, including lichens, carabid beetles, gastropods (slugs

Western Forest Strategy – AM Summary 18

and snails), and some bird species. Although songbirds associated with uncut forests are less abundant in retention harvesting, some were more abundant in group retention than dispersed retention. Bird communities in group retention blocks were more similar to uncut stands than clearcuts. There appears to be a positive relationship between patch size and persistence of some species after harvesting (i.e., larger patches are better).

2. How much retention is “best” (i.e, is there an optimal threshold of retention for maintaining most species? To date, the relationship between retention amount and organism abundance has been investigated using the response of birds. Often there is a 1:1 relationship – as retention levels increase, so do the abundances of forest species. However, the habitat attributes of the retention are important when examining the response of individual species. Huggard and Bunnell (2007) synthesized results for 69 bird species from over 50 studies with data comparing uncut forests and partial retention. As expected, there was a diverse range of responses to retention level. They found that many less sensitive species decreased substantially below 15 to 20% retention, and some more sensitive species, and those showing “soft thresholds” decline more abruptly below 35 to 40% retention. Some species showed an “above proportional” response to retention (abundance higher than % retention), while others showed a “below proportional” response (high levels of retention are required before any benefit is shown).

3. Are there major edge effects within aggregated retention? Lichen and moss species assumed to be dependent on old forests were sustained in older remnant patches without apparent edge effects. Preliminary findings suggest that relatively small patches may be sufficient to sustain them; however, these results are short term. There are edge effects that extend from the retention patch into the cutover area. For example, early edge effects are apparent in the abundance and richness of ectomycorrhizal fungi, and appear broadly similar to studies elsewhere. While the lack of an apparent edge effect within retained patches is encouraging, we expect some edge effects to become more evident with time. Both physical effects (e.g., windthrow) and biological effects take time to develop. Moreover, the character of edges changes with time. The preliminary results reveal the kinds of feedback possible, but it is longer-term edge effects that should inform management.

Vertebrate Species Monitoring

Landscape-Level summer bird surveys This study documents landscape-level trends in bird populations in forested landscapes. Project leaders are Wayne Campbell (Westcam Consulting) and Mike Preston (Jacques Whitford AXYS). A ninth season of monitoring was completed on summer bird transects in 2008 (a few transects have 10-year data). There are 52 active routes and approximately 2500 stations throughout coastal BC (30 routes on Vancouver Island [23 with 9 years of data], 16 on the Sunshine Coast [7 years] and 6 on Haida Gwaii/QCI [8 years]). The total dataset since inception of the study represents 115,605 individual bird detections of 131 species. The 10 most frequently observed species were American Robin, Winter Wren, Swainson’s Thrush, Townsend’s Warbler, Varied Thrush, Pacific Slope Flycatcher, Hammond’s Flycatcher, Warbling Vireo, Golden-crowned Kinglet and Chestnut-backed Chickadee. No monitoring occurred in 2009; however, we hope to obtain 10 year data on transects in 2010.

Western Forest Strategy – AM Summary 19

The intended purpose of the BBS method was to monitor a wide range of species and to provide insight into species-habitat relationships. A longer term goal is to provide estimates of species' trends, and potentially relate these to changes in habitat or structure availability. Preliminary habitat modeling was completed for Tree Farm License 39 using an earlier subset of our data (Vernier 2004). Subsequent analyses of trend are intended to serve as direct measures of change in species abundance, which indirectly may act as indicators of other biodiversity measures (e.g., changes in ecological processes, availability of specific habitat elements).

Trend estimates were made for select species and routes on Vancouver Island, and comparisons with provincial trends provided by the USGS were assessed. Winter Wren was the only species that showed consistent significant declines among all routes and regions that we evaluated. In the southwest and east-central regions of Vancouver Island, a significant negative decrease in abundance was observed for Golden-crowned Kinglet, Varied Thrush, and Chestnut-backed Chickadee. These forest-dependent species are identified as significant indicators of late-seral and old-growth forest conditions (Preston 2006). Hammond's Flycatcher was the only species among those tested that showed significant increases in abundance for both the southwest and east-central regions. Increases in species' abundance were most notable on northern parts of Vancouver Island, and included Townsend's Warbler, Swainson's Thrush, Warbling Vireo, Pacific-slope Flycatcher, and Wilson's Warbler. Among the 10 species we tested, 7 did not have significant trends at the provincial scale (Sauer et al. 2006). Of the remaining three species, two trends were consistent with our results.

An evaluation of regional versus provincial bird trends confirmed that differences are apparent. The most noteworthy observation was that among 11 species evaluated, eight had at least one significant regional trend but no corresponding significant provincial trend. Furthermore, Wilson's Warbler had a significant positive trend regionally, whereas it had a significant negative trend provincially. The consequence of mis-matched trends between provincial and regional assessments is likely to result in either ineffective or inefficient management. Our results suggest that trend estimates and species-habitat associations are likely to have substantial value for regional forest managers, and that continued trend monitoring will likely provide results that facilitate sustainable forest management, rather than hinder it.

In 2008 we recorded habitat data from 18 BBS routes on Vancouver Island and evaluated species-habitat relationships using a resource selection function and logistic regression. Among 15 species evaluated, significant response to forest type was observed for 12 species, and response to forest structure was observed for 9 species. Most importantly, among the significant habitat associations, six species showed response to habitat variables (shrub height, shrub cover, canopy cover) that are not well-represented in other habitat classification systems (i.e., Vegetation Resources Inventory or Forest Cover data). Our conclusion from this exercise was that a rapid habitat assessment using coarse-scale forest structure attributes can provide biologically meaningful results.

VRAM bird monitoring This project examines changes in bird communities at the experimental Variable Retention Adaptive Management (VRAM) sites that are the foundation of WFP’s monitoring program. The overall, long-term objective of this research, led by Dr. Ann Chan-McLeod at UBC, is to help determine how variable retention should be implemented in order to maximize its effectiveness in sustaining biodiversity. In 2008, mid-term (5-7 year post-harvest) breeding season bird response to different types of variable retention was assessed at 3 experimental sites: Pt. McNeill (group size), Tsitika (group %), and Horseshoe Lake (dispersed %). Preliminary models were also developed, combining short- and mid-term data, relating structure retention and habitat characteristics to bird diversity and abundance. Because of delayed FSP funding in

Western Forest Strategy – AM Summary 20

2009, field work at additional sites was not possible. Instead, additional data analysis was completed under the final year of FSP funding.

Preliminary results suggest that variable-retention harvesting was generally effective in retaining bird species that commonly inhabit mature forests; however, densities of these species were sometimes lower in the variable retention cutblocks than in the adjacent uncut forest. Variable retention cutblocks sustained higher densities of forest bird species than were detected in clearcuts. Overall results to date show that about one-third of individual bird species and all “guilds” (groups with similar habitat preference) were affected by retention level. Dispersed retention levels of 5% to 10% showed no significant overall difference in bird abundance from clearcuts, which favored shrub nesters and “open” habitat species. Group size appeared to have an impact on some species, with large groups (1 ha) preferred over small to medium-sized groups (0.25 to 0.5 ha), although results with retention type and level were highly variable.

Response of forest birds to group retention This study is a detailed examination of bird use of retention patches by Mike Preston, who completed his MSc thesis at Simon Fraser University in 2006. Results were published by Preston and Harestad in 2007 in Forest Ecology and Management. Species presence, abundance and habitat use was documented using point-count stations. Results show that the frequency of occurrence of common species from uncut stands is more similar after harvesting with group retention than with clearcutting – showing value in leaving patches. Data are from bird surveys over two seasons in 12 group retention stands, 12 clearcuts, and 12 uncut control stands, each containing five monitoring stations, and each surveyed three times each year. In total, 1,065 surveys were completed yielding 3,259 songbird observations. Results showed that all of the 18 most frequently detected species occurring in uncut stands were represented in group retention stands, but most (66%) occurred in lower abundance. Compared to clearcuts, group retention sites supported more species and, unlike clearcuts, were not dominated by Dark-eyed Junco and Winter Wren. The response of most forest-dependent species to percent retention appears relatively linear (i.e., greater abundance as retention increases).

Forest patch use by songbirds and woodpeckers Several key wildlife species or species groups are being monitored by Western Forest Products Inc. as part of the biodiversity effectiveness monitoring program in Englewood Forest Operation’s Tree Farm License (TFL) 37 (Manning and Cooper 2007). Monitoring of songbirds and woodpeckers in various ways was initiated in 2003 and has continued through 2007. Specific project objectives in 2007 were: 1) Monitor songbirds and woodpeckers in cutblocks which have treed retention patches (patches were previously surveyed for forest structure); 2) Map and compare bird locations in relation to patch locations in cutblocks; 3) Correlate songbird occurrence with retention patch characteristics (size, location, forest structure).

Point count census station were established and flagged in 32 forested retention patches in 11 cutblocks. Additionally, points were placed in two unharvested old forest control/reference sites. A total of 956 detections of 49 species of birds occurred during surveys in 2007. Of these, 328 detections of 32 species (or ~34% of detections) occurred within retention patches. Approximately 60% (577/956) of total detections occurred outside of retention patches (either in harvested areas or adjacent forest), while 5% (51/956) occurred at the two old growth reference sites.

Based on the survey data collected in 2007, the following implications are apparent concerning forest interior birds and retention patch characteristics:

Western Forest Strategy – AM Summary 21

• Patches >3.4 ha had species richness approximately the same as old forest reference sites. • Patches <3.4 ha had substantially lower species richness than reference sites. • Fewer birds occupied small patches later in the breeding season, suggesting reduced temporal occupancy of small patches. • Forest interior bird species detected in patches <3.4 ha (and particularly under 1.0 ha) may not be breeding, or not breeding successfully, in the smaller patches.

Unfortunately, so few forest interior species were present in small patches, that correlation analyses between bird occurrence/abundance and the forest structure data collected from these patches in 2005, was neither meaningful nor possible.

Thus, based on the results of this study, it appears that patches <3.4 ha in size have substantially reduced occupancy and/or use by breeding forest interior bird species, and may not provide adequate habitat for some of these species. At this time, however, because of limited sample sizes, rigorous conclusions about optimum patch size are still not possible – further research in this area is needed. Nevertheless, the 2007 data do provide notable trends which can be directly applied to current forest management practices.

Aquatic breeding amphibians: harvesting impacts and habitat identification This project includes two components: 1) a study of the effects of harvesting and buffers on amphibian use of small wetlands and 2) development of a field card for identification and ranking of small ponds for amphibian habitat. The field study is located on three cutblocks within the private lands of Island Timberlands. Different buffers widths were assigned to small wetlands to test the efficacy of retention patches for maintaining breeding populations of amphibians. WFP and IT shared the cost of the study through 2007; future monitoring will be done as funding is available.

In 2007, the second season of post-harvest monitoring occurred on two sites and continued for a third season on a third site. Unfortunately, loss of buffer replication occurred on two sites during logging operations; however, these sites can be monitored for harvesting impacts. Four amphibian species were found breeding in small ponds: long-toed salamander, rough-skinned newt, Pacific chorus frog and red-legged frog.

Two and three years after harvesting, the four amphibian species have continued breeding at small wetlands at each site. Frogs utilized more wetlands for breeding post harvest than during pre-harvest year(s). Wetlands with no canopy cover appeared to have a greater influx of breeding post harvest than wetlands with retention. Reduced hydroperiod has not been a threat to amphibian breeding success at these sites as wetlands have retained water longer post harvest than in pre-harvest year(s) as a result of deeper water and slower drying rates. In-pond conditions have been suitable for reproductive success, as metamorphosis was observed each year at the majority of wetlands where breeding was confirmed.

The results of this study to date have been as expected based on observations for amphibians elsewhere. Because many frog species demonstrate a preference for low to mid-level canopy cover conditions for breeding, we expected that some species might take advantage of newly harvested sites to exploit ‘new’ wetlands. For example, one of our sites had numerous observations of Red-legged Frog adults before harvesting but none of the wetlands were used for breeding. Since harvesting, Red-legged Frog breeding has been observed at four wetlands. These are initial results and continued monitoring is required to determine whether conditions will remain suitable for breeding once green-up occurs and the forest canopy recovers.

Western Forest Strategy – AM Summary 22

The field card for identification of important amphibian habitat by forestry planners was completed with funding by the BC Forest Science Program. A flow chart helps the user identify and rank small wetlands for their potential as habitat for amphibians. An accompanying card provides recommendations for practices that protect habitat features. A summary describing the field card and its use was published: Wind, E. and W.J. Beese. 2008. Little known and little understood: Development of a small wetland assessment field card to identify potential breeding habitat for amphibians. BC Journal of Ecosystems and Management 9(1):47–49.

Invertebrate Species Monitoring

Effects of retention on carabid beetles This study examined the potential use of carabid (ground dwelling) beetles as indicators of forest conditions and biological diversity under various stand-level retention strategies. Initial pilot studies by Dr. Isobel Pearsall began in 2002 to fine-tune pitfall trap transect sampling methods and to characterize diversity. A study of post-harvest ground beetle populations at VRAM sites began in 2005-2006 to examine the effect of retention group size at the Pt. McNeill and Klanawa sites. A three year project to obtain 4- to 6-year post-harvest data at the remaining 7 VRAM sites began in 2007 with WFP funding, and continued in 2008 and 2009 with FSP support. In 2008, sampling occurred at Goat Island (group %), Lewis Lake (riparian) and Moakwa (riparian). In 2009, sampling occurred at the Memekay (group removal) and Hoodoo (group %) sites.

Studies at the two group size VRAM sites found a response of carabid beetles to retention patch size, with higher catches per trap of “forest specialists” in larger patches than in smaller patches. This response was seen in both old growth and 2nd growth sites, and at both south- western and northern Vancouver Island sites. Small patches of 2nd growth were particularly poor at retaining forest specialist beetles. At this time, the direction of carabid species succession is uncertain, since we do not know the possible extent of competitive interactions and whether the size of patches is adequate for long-term survival of beetles.

In 2008, 16 carabid beetle species (3288 individuals) were collected at the Moakwa site, 15 species (3230 individuals) at Lewis Lake site, and 15 species (3785 individuals) at Goat Island. A total of 20 carabid species (10,303 individuals) were trapped overall. Interim findings from the three sites monitored in 2008 show that there appeared to be a clear separation of the unlogged “control” and patch communities, and the matrix (harvested areas surrounding patches) and clearcut communities. Edge communities were more similar to control communities, while communities from traps located just 5-10m from the edge of patches had a more similar composition to the matrix communities. Thus, there does appear to be some indication of “typical” interior, edge and matrix communities at these VRAM sites. Two species were strong indicators for patch conditions.

Diversity of carabids was highest in the clearcuts and lowest in the unlogged controls. This is a result of the dominance of the forest specialist Scaphinotus angusticollis in controls, while clearcuts and matrix sites tend to receive an influx of small, winged species that are open- habitat specialists. Although there were some differences in terms of species composition among the old-growth site at Moakwa and the other two sites, there did not appear to be a particular “riparian” community associated with the retention buffers. Group retention patches at Goat Island were generally not effective at retention of the original second growth communities of carabid beetles 4 years post-harvest, most likely due to the high levels of windthrow at this site. Final results for the remaining VRAM sites sampled in 2009 will be completed in April 2010.

Western Forest Strategy – AM Summary 23

Terrestrial gastropods as biodiversity indicators This study examined the potential use of gastropods (slugs and snails) as indicators of forest conditions and biological diversity under various stand-level retention strategies. Intensive pre- disturbance surveys of slugs and snails using litter sampling and artificial cover objects were completed for 6 VRAM sites from 2001 to 2003 by Lennart Sopuck and Dr. Kristiina Ovaska of Biolinx Environmental Research. Twelve to 16 species of gastropods were found on each site, comprised of small litter snails, large snails, carnivorous snails and slugs. Small litter snails accounted for the most species on all sites. Post-harvest surveys were completed on the 6 sites from 2005 to 2007. A final report, draft publications and an operational summary were completed in March 2008.

Results for the six experimental sites revealed that 10 of 12 species of gastropods tested showed a significant treatment effect in pre- and post-logging comparisons over the short term (2 – 4 years after logging). The effects were negative for six species, positive for two species, and two showed mixed responses (positive or negative) depending on the site. Significant treatment effects, all negative, were also found for small snails as a group and for species richness at most sites, but not for the overall species diversity, as measured by the Shannon- Wiener index. Three common, ecologically important gastropods (Ariolimax columbianus, Prophysaon foliolatum, and Vespericola columbianus) benefited from VR-treatments at some sites. VR-dispersed treatment with 30% retention level was better at maintaining gastropod abundance and species richness than were treatments with lower retention levels, but it was not equivalent to the uncut control. Similarly, large retention groups (0.8 – 1.2 ha) were more effective than were smaller groups (0.2 – 0.5 ha; < 0.2 ha) or clearcut. VR-group retention with small groups (0.2 – 0.5 ha) was similar to 20 ha clearcuts in maintaining sensitive gastropods regardless of the overall level of retention (10%, 20%, or 30%) in the cutblock.

The study authors recommend the following practices for maintaining conditions suitable for gastropods and associated forest floor organisms: 1) retention groups should be as large as possible (0.8 – 1.2 ha or more), especially in even-aged second growth stands; 2) If dispersed single trees or small clumps of trees are used, a retention level of at least 30% is recommended; 3) Strategies that conserve moisture and protect sensitive riparian zones are desirable, including anchoring retention groups on wetlands or moist depressions.

Variable retention, small streams and invertebrates A large, integrated experiment examining variable retention on small headwater streams was started in 2001. Two of the company’s VRAM sites (Lewis Lake near Powell River, and Moakwa Creek near Sayward) were designed for this study. A description of the full study design and objectives is given in Appendix 5. Unfortunately, funding cuts terminated the project in 2003; however, considerable pre-harvest data was collected at each site and harvesting was carried out according to the study design. Stream monitoring continued with limited funding until 2008. Small streams were also studied at a third VRAM site (Tsitika, group retention level) as part of a PhD thesis at UBC; unfortunately, the student did not finish the program. Samples of stream invertebrates and other data from this project were used as the basis for a BSc thesis by Amber Paulson at Malaspina University-College (now Vancouver Island University) in Nanaimo.

Paulson (2007) compared the emergence of aquatic true flies (Diptera) from small streams with uncut, clearcut and retention treatments. Total abundance and taxa richness was similar for all three treatment sites. Uncut sites were found to have lower diversity and a different community structure than clearcut and retention sites. Clearcut and retention sites were found to have a larger abundance of predators than uncut sites. Uncut sites had a higher average abundance of fungus gnats than clearcut or retention sites. This study showed that clearcut logging affected

Western Forest Strategy – AM Summary 24

aquatic Dipteran communities one and two years after harvest by altering the community structure, and increasing diversity and predator abundance. Aquatic Dipteran communities sampled from the retention site were found to be similar to those communities sampled from the clearcut sites, and not similar to those sampled from the uncut sites; however, the downstream location, lack of replication and windthrow associated with the retention stream limit the reliability of this conclusion. The implications of this short-term study for stream recovery and long-term effects of harvesting on the complete aquatic invertebrate community are uncertain.

Vegetation Species Monitoring

Ectomycorrhizal fungi and retention A study was established in 2005 to examine the abundance of ectomycorrhizal (EM) fungi on tree seedlings in relation to dispersed trees and cutblock edges under the direction of Dr. Tony Trofymow with the Canadian Forest Service. Transects were established at the dispersed VRAM site (Horseshoe Lake) in Stillwater Timberlands near Powell River, BC. A chronosequence of sites was also studied at the Northwest Bay operation of Island Timberlands to examine persistence of mycorrhizae in different ages of forest in proximity to mature timber edges. Results showed that the abundance of EM fungi decreased with distance from forest edges, including group retention. It appears that isolated trees maintain a certain proportion of mycorrhizal communities characteristic of mature forests; however, the role of single live trees in providing viable long-term support for survival and growth of EM fungi is uncertain. Findings from the dispersed retention study were published: Outerbridge, R.A. and J.A. Trofymow. 2009. Forest management and maintenance of ectomycorrhizae: A case study of green tree retention in south-coastal British Columbia. BC Journal of Ecosystems and Management 10(2):59–80.

Bryophytes and retention This project investigated patterns of vegetation abundance and species richness in regenerating and mature CDF and CDF/CWHxm transitional forest. Specific objectives were to (1) inventory vascular plants and bryophytes, (2) investigate small-scale bryophyte-habitat relationships, (3) characterize vegetation patterns in different age classes (5-year old, 25-year old, 55-year old, and 90+ year old), and (4) characterize vegetation patterns at regenerating forest edges (i.e. at the interface of each of 5-year old, 25-year old, and 55-year old stands with 90+ year old forest). Patterns in reproductive status, and influence of aspect were also examined.

Three sites were sampled per regenerating forest age class (5-year old, 25-year old, 55-year old) and 90+ year old forest interface. For each site, vegetation was sampled in four 15m x 22.2m plots. Three plots were located in regenerating forest, with microplot sampling along nested transects at 5m & 15m, 25m & 35m, and 60m & 70m from 90+ year old forest edge (two transects per plot). The fourth plot was located in the adjacent mature forest stand (nested transects at 60m & 70m).

Within each plot, six 1m x 1m microplots were sampled for vascular plants. Bryophytes were sampled on logs, tree bases, and on the forest floor (six 0.3 m x 0.1m microplots for each substratum-type). Plots were surveyed for additional vascular plant and bryophyte species. The reproductive status of each species recorded in plots was qualitatively observed and recorded. To investigate age class, only plots associated with 60m & 70m transects were compared. To investigate edge effect, all plots were compared.

In total, 158 plant species were identified within the sample plots: 90 vascular plants (35 trees and shrubs, 30 herbs, 25 graminoids including grasses, sedges, and rushes), and 68 bryophytes (53 mosses and 15 liverworts). No rare taxa were recorded, although two species

Western Forest Strategy – AM Summary 25

listed as of interest to Weyerhaeuser for tracking purposes (Bunnell et al. 2003) were observed. Many rare and sensitive species in the CDF zone occur in the driest portion of the landscape (e.g. south-facing rock outcrops). These habitats may provide ideal anchors for mature forest remnants in CDF zone forests.

Patterns of bryophyte abundance and species richness reflected substratum-type (logs, tree bases and terrestrial environments), log decay level, and tree-type (coniferous-vs. deciduous). Bryophyte species that demonstrated an affinity for particular substratum-types or qualities are discussed. Most bryophyte species demonstrated a preference for moderately to well-decayed logs, and coniferous tree bases; few species were most prominent in forest floor habitats.

Age class influenced patterns of vegetation cover and species richness; these qualities were related to stand variables (canopy cover, coarse woody debris levels). Overall, few vascular plant species demonstrated a sensitivity to mature-forest conditions (i.e. were only found in 90+ stands). The saprophyte Boschniakia hookeri may be a useful indicator of mature forest in the CDF zone. In contrast, many bryophyte species were documented only in 90+ year old stands, or increased in prominence in relation to stand age (e.g. Dicranum fuscescens, Hypnum circinale, Ptilidium californicum, and Scapania bolanderi).

The influence of edge proximity was most notable in 5-year old stands, for the variables canopy cover and vascular plant coverage, richness, and reproductive frequency. In contrast, the relationship between edge proximity and bryophyte coverage (overall, and on substratum-types) was most notable in 25-year old stands. Trends in bryophyte richness with edge distance varied between substratum-types, but was consistent for all age classes for bryophytes sampled on logs (i.e. richness declined with increasing distance from mature edge).

No vascular plant species demonstrated a clear relationship with mature-edge proximity in terms of coverage or reproductive status. In contrast, several bryophytes (Scapania bolanderi, Ptilidium californicum, Dicranum fuscescens, and Hylocomium splendens) were more prominent in regenerative plots nearest to mature forest edges. For these species, adjacent mature forest reserves may act as “lifeboats” for recolonizing populations. The liverwort Scapania bolanderi was found to be an ideal indicator of mature forest conditions; populations were larger in older forests, and coverage values reflected the proximity of older forests.

Aspect was briefly investigated as a factor impacting vegetation patterns in sample areas. Edge and slope aspect influenced the coverage and richness of herbs and bryophytes in both regenerating 5-year old forests and adjacent mature forest stands. Many species in regenerating forests (particularly bryophytes) may depend on propagule inputs from adjacent mature forest habitat. As such, relationships between aspect and species richness in remnant patches (and associated regenerating forest) should be investigated more thoroughly.

Canopy epiphytes and retention Lichens occurring in forest canopies of coastal forests were piloted to determine best field sampling protocols and baseline information on species abundance and richness. This project was specifically focused on sampling the Tsitika and Stillwater Variable Retention Adaptive Management (VRAM) Experiments. Lichens occurring in coastal forest take a long time to establish and are weak dispensers from forest edges. It is uncertain whether variable retention harvest areas adequately maintain canopy lichen populations and if they improve the rate of colonization of the harvested matrix. These studies provided baseline information on canopy lichen abundance and richness under an old growth and older second growth forest condition. This data will be used to compare the effects of varying retention levels and types on persistence and re-occupancy time.

Western Forest Strategy – AM Summary 26

Montane Alternative Silvicultural Systems (MASS) – vegetation studies The MASS study compared clearcut, patch cut, dispersed (green tree) retention and shelterwood systems for managing coastal montane ecosystems.

Analysis of permanent understory vegetation plots over 15 years showed that: • The shelterwood maintained the greatest diversity of understory trees, shrubs and bryophytes compared to the other systems, and maintained the greatest similarity to the old growth control in terms of overall species and life form composition. • Vegetation reached at least two-thirds of pre-harvest cover after 5 years in all treatments and had surpassed pre-harvest cover after 10 years. • Herbaceous colonizers accounted for most of the rapid increase in vegetation cover in years 3 through 10; this was most pronounced in the clearcut, patch cut and green tree treatments. • The post-harvest decline in species associated with undisturbed forests and increase in early seral species was least pronounced in the shelterwood treatment. Species gains exceeded losses on treatment plots; however, bryophytes and herbs that prefer moist, shaded habitats generally decreased after harvesting.

Other biodiversity studies at MASS include birds and canopy insects.

Species at Risk Monitoring

Northern goshawk nest territory monitoring The coastal subspecies of Northern Goshawk, the (Accipiter gentilis laingi), is an uncommon forest raptor that is designated as Threatened by the Committee on the Status of Endangered Wildlife in Canada (COSEWIC). In WFP’s TFL 37 on north-central Vancouver Island, 53 nest trees in 13 confirmed goshawk territories and 2 suspected goshawk territories have been found (1994-2006). In 2006, TFL 37 was transferred from Canadian Forest Products (Canfor) Ltd. to WFP. Previous to 2006, Canfor had committed to managing habitat for this species through an adaptive management strategy involving approved Wildlife Habitat Areas (WHAs), which vary in size from approximately 135-538 ha. Ten goshawk WHAs were established in TFL 37 by the BC government in March 2003.

In June and July 2008, surveys to determine the breeding activity status were conducted in 12 known goshawk territories in TFL 37. All known territories were visited at least twice during the survey period (once during the nestling period and once during the fledgling period) to determine the activity status and number of fledglings in each nesting territory. Six goshawk detections occurred during surveys in 2008, which resulted in four active nests being located. Overall during 2008, 50.0% (6/12) of surveyed territories were occupied; this was consistent with the long-term average (1995-2008) of 48.0% (n=51/113) occupancy for known nest territories in this area over a 14 year period (E. McClaren, BC MoE, unpublished data). Average productivity for 2008 was 2.75 fledglings per nest, which is the highest productivity in TFL 37 since monitoring began in 1994. The mean long-term productivity rate (1994-2008) for known nest territories in TFL 37 was 1.61 young fledged/active nest/year.

Including the 2008 surveys, there are now 64 known nest trees in TFL 37 that have been documented or are thought to be those of goshawks. Analysis of prey species showed a positive correlation between the number of red squirrel detections in a given year and the number of known goshawk territories that were occupied.

Western Forest Strategy – AM Summary 27

Old-growth specklebelly lichen survey Old-growth specklebelly (Pseudocyphellaria rainierensis) is a lichen found on the west coast of Canada and the U.S. from Oregon to southeast Alaska. Current known locations are primarily associated with late-seral or old-growth forests. Old-growth specklebelly (OGS) lichen is currently on Schedule 3 (Special Concern) of the Species at Risk Act (SARA) with a re- assessment expected in April 2010. Collections in recent years on northern Vancouver Island (NVI) suggested that this lichen may be more common than previously thought in this portion of its range, with important implications for forest management. The objectives of this project were to obtain baseline data on the relative abundance and distribution of OGS lichen in several WFP TFLs (6, 19, 25, 39, 44) and adjacent parks that would provide improved data to inform COSEWIC designation and forest management strategies.

In 2008, surveys were completed in seven watersheds within two TFLs on NVI. OGS lichen was found in all watersheds and on 58% of the sites surveyed (29/50). OGS lichen appears to be more common in TFL19 (72% of sites) than TFL6 (33%), although some of the sites with the highest abundance were found in the Cayuse valley of TFL6. The frequency in the TFL19 Gold River operation seems at least partially due to the greater proportion of high elevation forests with a component of yellow-cedar than in the Jeune Landing and Holberg operations where the topography is more subdued.

In 2009, additional surveys were completed in three watersheds on NVI (TFL39), several watersheds within TFL25, TFL44 and adjacent parks on southern Vancouver Island (SVI) and in the Stafford watershed (TFL25) on the central mainland coast (CMC). In TFL39, OGS lichen was found in all watersheds and on 65% of the sites surveyed (13/20). Overall, we found OGS lichen on 60% of the 70 plots established over two years NVI. No OGS was found within the 16 stands surveyed in TFL25 on the southern tip of VI, despite the presence of old-growth habitat. In TFL44 and adjacent Carmanah-Walbran Provincial Park on SVI, OGS lichen was found on only three sites in one of the four major watersheds surveyed (outside the park), for a success rate of 15% (3/36 sites). In contrast, we found OGS lichen throughout the Stafford watershed on the CMC. We collected OGS on 92% of sites (12/13) following the survey protocol and found it on 11 additional sites with visual surveys. Including four sites where OGS lichen was sampled during other field trips on VI, we have documented 72 new locations for this species, for which only 16 previous locations had been documented since its first discovery in Canada (BC) in 1950.

OGS lichen was found across a wide range of ecosystems as classified by biogeoclimatic site series (i.e., plant associations). Overall, we found OGS in 20 of 30 (67%) of the site series represented in sample plots. The majority of sites were in the montane CWHvm2 variant and at upper elevations in the CWHvm1 because surveys focused on forests with a component of yellow-cedar (Yc). For the CMC, we found OGS in all 8 of the site series surveyed across a range of elevations and biogeoclimatic units (150m to 780m; CWHvm1, vm2, MH). NVI surveys found OGS as low as 185m elevation in the White River on productive, valley bottom sites without yellow-cedar and on half of the NVI sites (4/8) in the Mountain Hemlock (MH) zone. Across all survey locations, we found OGS lichen on sites ranging in elevation from 150m to 980m and spanning the full range of moisture and nutrient regimes. We found OGS on five sites representing four site series in the CWHmm2 variant in the Memekay watershed. We also discovered OGS on three sites on SVI (south of Pt. Alberni) representing three site series in the CWHvm1. The latter two finds are significant because OGS has not been found in the CWHmm2 variant or on southern Vancouver Island since the July 1950 collection by Vladimir Krajina.

Western Forest Strategy – AM Summary 28

Although widely distributed, OGS lichen achieves its greatest abundance on nutrient rich middle to lower slopes or microsites. Prime habitat for OGS is a moderate to productive CWHvm2 site with Yc or a productive CWH vm1/vm2 site with amabilis fir. Surveys confirmed that a preferred habitat is a lower-canopy amabilis fir within the canopy drip of yellow-cedar; however, we found OGS on a variety of host species: western hemlock, mountain hemlock, yellow-cedar, western yew and western redcedar as well as dead amabilis fir.

Where OGS was found, our average search time was 12 minutes with a two-person crew. Colonies varied from a few small individuals on branches to large patches on old amabilis fir stems that were quite vigorous. At many locations we spotted OGS from the roadside, or found it while examining the first suitable tree in the survey area. Two occurrences of OGS adjacent to the White River main road, which was built over 30 years ago, suggest that OGS is tolerant of edge influence. We collected over 86 specimens with accompanying photos, habitat descriptions, site descriptions and vegetation cover estimates.

Results of these surveys show that old-growth specklebelly lichen is widely distributed on at least the portion of NVI and the CMC surveyed. Its presence on SVI was documented from several sites, including photo evidence from ecosystem mapping plots in Clayoquot Sound where significant old-growth habitat is protected. Existing protected areas and biodiversity conservation strategies in BC government regulations and certification systems provide landscape-level reserves and stand-level retention that should maintain significant old-growth habitat to sustain this species in coastal BC. Discovery of abundant OGS lichen on the central mainland coast, managed under Ecosystem Based Management with a high proportion of protected area, is particularly encouraging for ensuring the conservation of this species and other old-growth associated lichens.

Review and Summary of Silviculture Monitoring and Research

The objectives of silviculture monitoring are to: • Assess potential negative impacts on regeneration and forest growth from edge effects (i.e., light, water, nutrients), dwarf mistletoe and root disease; • Determine the amount and severity of wind damage in recent harvested areas; • Develop strategies and predictive tools for planning, cutblock design and silvicultural treatments to minimize negative impacts of retention on timber value, forest growth and site productivity.

The key questions relating to growth and yield concern how different levels and spatial patterns of retention affect growth of different tree species. Wind damage, hemlock dwarf mistletoe and root disease have been identified as potential forest health problems that could be increased through variable retention. Although these factors are part of natural disturbances, they could lead to negative impacts on timber production. Losses from wind and pathogens are unavoidable; however, the risks and economic impacts can be managed through planned salvage or retention of downed wood when it meets habitat objectives.

Guidelines for minimizing the spread of dwarf mistletoe and root rot in the design of variable retention were written for the company’s “Silvicultural Prescriptions for VR” document. Guidelines were based on recommendations in BC MFR guidebooks. Approaches include the removal of infected trees along timber edges and planting non-susceptible species. Implementation monitoring found that silvicultural prescriptions did a good job of identifying root rot and mistletoe hazards and prescribing management strategies. Monitoring dwarf mistletoe at the MASS research site found no significant infections on regenerating hemlock; this result is

Western Forest Strategy – AM Summary 29

applicable to montane or higher elevation sites. Interfor is collaborating with pathology expert Dr. John Muir to develop a mistletoe monitoring program for retention harvesting that will assess the potential impacts on regenerating stands. Currently, WFP has no separate monitoring program for root rot and mistletoe; however, these pathogens are identified as part of growth and yield monitoring.

Variable Retention Windthrow Monitoring

Improving our ability to avoid excessive wind damage is one of the important elements of implementation and monitoring. During 2000, we developed a strategy for wind hazard prediction and monitoring. Dr. Steve Mitchell and graduate students at UBC developed a wind hazard model and maps for operational planning. The work built upon a model developed for WFP on northern Vancouver Island. Hazard maps were produced for the company’s crown- land operations. The model uses topographic data, stand attributes and historical windthrow occurrence to predict windthrow risk to forest edges. Maps can be produced for different levels of damage and risk.

The windthrow monitoring strategy consists of measuring wind damage on operational VR cutblocks and VRAM experiments, building a database covering our entire tenure, analyzing data and communicating the results and trends to operations. Through contracts with Terry Rollerson (Golder Associates) and Colin Peters, we developed a cutblock-level windthrow monitoring program and database. After a two-year pilot program to assess methods in operations with the highest windthrow incidence, monitoring over the past 6 years has assessed damage to retention groups, dispersed trees, riparian buffers and cutblock edges on a random sample of blocks in each operation on a biennial cycle. VRAM sites are assessed in year one, three and five post-harvest. For external edges and larger groups, visual estimates of the amount of windthrow and depth of penetration are recorded the first 25 metres into a stand edge. Both the hazard mapping and monitoring programs were supported by the provincial government through FRBC and FIA funding programs.

Windthrow monitoring was carried out over eight years (2001-2008). We chose a random selection of operational cutblocks in several operations each year, initially in legacy Weyerhaeuser operations. After the merger with WFP, monitoring was expanded in the final two years to four operations on north-western Vancouver Island: Holberg, Jeune Landing, Gold River and Nootka Sound. We also included nine Variable Retention Adaptive Management (VRAM) experimental areas in the monitoring schedule. The distribution of monitoring sites on representative sites across Vancouver Island (VI), the southern BC mainland coast and Haida Gwaii (Queen Charlotte Islands) facilitated evaluation of coast-wide variation in windthrow associated with variable retention (VR) harvesting practices, especially the retention silvicultural system. The wide geographic extent, however, complicated data analysis due to the high spatial variability in terrain conditions and wind regimes among the various study areas.

The overall project objectives were: • Document the amount, spatial distribution and regional differences in wind damage associated with VR. • Identify the qualitative and quantitative factors associated with VR wind damage including both environmental factors and treatment effects. • Identify specific management options to control wind damage associated with VR. • Develop field indices and decision-making tools to enhance windthrow hazard assessments. • Communicate the results to operations staff.

Western Forest Strategy – AM Summary 30

The project database contains 4648 plots within 172 harvested cutblocks. Plots represent nearly 366 km of external cutblock boundaries, 26 km of large patch edges, 197 hectares of small retention patches and 50 km of riparian and other strip edges.

The study showed definite regional differences in wind damage4 for cutblocks that have experienced at least two winter wind seasons. The average percentage of wind damage along external cutblock edges varied from an average of 11% on southeast Vancouver Island (South Island or Island Timberlands) to 25% on northwest VI near Quatsino Sound (Jeune Landing FO), with an overall average of 16±0.2%5 across all areas. There were similar regional differences in wind damage for retained patches. The average wind damage along the edges of larger patches was 16% in Stillwater and South Island and 45% in the Queen Charlotte FO with an overall average of 24±0.6%. The average wind damage in small patches (i.e., ≤ 1 hectare in area) ranged from 20% in South Island and 21% in Gold River to 45% in Queen Charlotte and Mid Island FOs with an overall average of 39±0.5%. For the edges of strips of retained timber, the wind damage amounts ranged from 15% on southeastern VI (South Island) to 38% on northern VI (Jeune Landing and Port McNeill), with an overall average of 31±0.6%.

Analyses of the data from the various strata (cutblock edges, patches, strips) suggest the following general relationships: • Windward edges along external cutblock boundaries, large patches and retained strips are more vulnerable to wind damage than other boundary exposures. • Wind damage tends to increase with increasing fetch distance to the edges of block boundaries, large patches, small patches and retained strips. • There is a strong relationship between slope position and the amount of wind damage. Topographically exposed locations such as ridge crests and upper slopes tend to experience more wind damage. This trend matches a general, but weak trend of increasing wind damage with increasing elevation above sea level. There is some indication that steeper slopes may be more vulnerable to wind damage. Slope angle, however, tends to increase with increasing elevation so the relationship between wind damage and slope angle may be correlated with exposure. • For external edges, large patch edges, smaller patches and retained strips, the amount of wind damage increases with increasing stand height, but also with increasing rooting depth. • In general, external cutblock edges and the edges of retained strips are more vulnerable to wind damage when they occur along the edges of gullies or stream escarpments than in other topographic locations. Providing setbacks from the edges of these topographic features tends to reduce the amount of wind damage and/or reduces the likelihood that wind damage will penetrate into these features. • Stands that appear to have established after previous windthrow events appear to be more vulnerable to wind damage than stands originating in other ways (e.g., wildfire, harvesting or without recent stand-replacing disturbance, as is the case for old growth stands). There is some indication in some areas that second growth stands may be more vulnerable to wind damage than old growth stands. In contrast to the above, stands categorized as multi-storied appear to be more vulnerable to wind damage than uniform, single-storied stands. • Windthrow penetration tends to increase as percent wind damage increases. Penetration distances increase with increasing exposure to wind, being least on lee

4 Total wind damage % = % windthrow (trees uprooted) + % stem-break + % leaning 5 Standard error of the mean.

Western Forest Strategy – AM Summary 31

boundaries and greatest along windward boundaries. A similar relationship occurs with cumulative fetch distance, with penetration increasing as fetch distance increases. Windthrow penetration increases with increasing stand height and increasing rooting depth and windthrow penetration distances are greatest along the edges of gullies and steep stream escarpments. • Among the major coastal coniferous tree species, western hemlock and amabilis fir are the most vulnerable to wind damage. Western redcedar and yellow-cedar are generally less susceptible to damage. Douglas-fir appears to be the most windfirm coastal conifer (with the possible exceptions of lodgepole pine and white pine). Red alder and bigleaf maple are also quite vulnerable to damage on external cutblock edges. There are greater differences among species and greater variation in wind damage for more exposed conditions (e.g. small patches, strips) than for large patches and external cutblock edges.

Management implications from our findings include: • In situations where the hazard of wind damage is high (i.e., considering geography, site and stand conditions) use large patches and wide retention strips rather than small patches, narrow strips or dispersed retention. Where the hazard of wind damage is very high, clearcut with reserves may be more appropriate than the retention system for maintaining stand-level retention. • For gully edges and stream escarpments, set back boundaries 10 to 15 metres from the windward edges to reduce the potential for windthrow penetration into gullies and across streams. The data suggest that setbacks of even a few metres will experience lower wind damage rates than boundaries located directly along the edges of gullies and stream escarpments. • When feasible, top and prune trees along edges where damage is likely to compromise management objectives. Although monitoring findings were inconclusive due to low sample size and high variability, data from other studies suggest that windthrow can be reduced along windward and windward diagonal boundaries with a treatment depth of at least 15 to 20 metres. • Where possible avoid locating windward boundaries of VR patches or retention strips in tall timber.

Growth and Yield — Edge Regeneration Studies

Several studies directed by Dr. Nick Smith examine the influence of forest edges (either adjacent stands or retention patches) on seedling growth as well as the growth of retained trees. Sites include experimental VRAM areas, supplemental AM comparison sites and operational cutblocks. Two current FSP-supported projects provided funding for field measurements, analysis and modelling. Additional field work was supported by the FIA Land-Based Investment Program in 2009. Most sites are being measured on a 1, 3, 5 and 8 year cycle. In addition to planted tree transects, natural regeneration and retained trees are also measured.

The growth and yield program has also established permanent plots within the 9 VRAM experimental sites. In each treatment, "sector plots" were established to collect growth data for both the leave trees and planted and natural regenerating trees (Iles and Smith 2006). This novel sampling approach collects information in an unbiased fashion from retained groups and surrounding areas, correctly accounting for edge effects. This project has continued to develop the sector sampling approach. A paper entitled “Sector sampling: some statistical considerations”, N.J. Smith, K. Iles and K. Raynor was published in Forest Science in 2008. The

Western Forest Strategy – AM Summary 32

latest work extends sector sampling to fixed area plots such as circular plots as is useful for measuring natural regeneration.

In order to increase the sample size for growth and yield data and cover more stand conditions than are possible in the VRAM sites, we have established supplemental comparison sites with smaller size requirements, including mixed group-dispersed retention. These sites are operational cutblocks, but treatments are allocated randomly to portions of the blocks. We have also established permanent growth and yield plots in each treatment replicate at the Montane Alternative Silvicultural Systems (MASS) site, which are now 15 years post-harvest.

A spatially explicit, interacting grid model is under development (FORGE) that tracks growing season light and moisture for fir and hemlock. The model utilizes permanent sample plot data and produces results consistent with the field data. The modelling effort includes collaboration with Dr. Andy Black (UBC) on microclimate components. We have climate station data from experimental sites that are being fit to the growth model, including transects away from large and small patches and individual trees. Photosynthetically active radiation, temperature and moisture are being measured along each transect for the growing season (early, mid and late summer). Fish-eye imagery along transects is used to quantify canopy cover.

Findings to date suggest that most of the impact of group retention on tree growth is restricted to 10m from the edge but is noticeable in all directions for all species examined. The results were highly correlated with surficial moisture. Impacts when spread across the cutblock, however, were small. There was no difference in impacts with azimuth due to high microsite variability for young trees. Greater seedling impacts were noted for small group removal sites, rather than external cutblock edges or group retention.

A draft paper for journal publication was prepared in 2009 for the dispersed VRAM site and two other dispersed retention experimental sites. All sites were measured for planted Douglas-fir tree growth and survival five-to-six years after establishment. The dispersed trees were predominantly large diameter (60cm+) Douglas-fir left with a range of 0% to 30% of the original basal area (0 m2/ha to 14 m2/ha). Two sites had 0%, 5% and 15% retention, while one site had 0%, 5%, 10% and 30% retention. The trees were measured in sector plots established to randomly sample the range of microsites in each treatment. There was no statistical difference between height and basal diameter growth between the retention treatments over the measurement period, except for basal diameter growth at the 30% retention level. Young tree growth equations were fitted to data from one site as part of the FORGE model examining the impacts of light availability on growth as affected by different levels of dispersed retention. The small impact of low levels of retention below 15% suggests that higher levels of retention are needed to significantly affect Douglas-fir planted tree growth.

Montane Alternative Silvicultural Systems (MASS) – Regeneration Studies

This study was established in 1992 to examine alternative approaches for managing high elevation forests. MASS is located southwest of Campbell River in Island Timberlands’ (formerly Weyerhaeuser’s) Iron River Operation in the CWHmm2 variant. Four silvicultural systems are compared: clearcutting, small patch cutting (1.5 – 2.0 ha), dispersed (green tree) retention, and a uniform shelterwood with 25% basal area retention. Adjacent to these treatments is 20-ha old- growth baseline monitoring reserve. In 2008, windthrow salvage and patch cut harvesting was completed by Island Timberlands. Two patches approximately 0.8 ha remain in each patch cut replicate.

Western Forest Strategy – AM Summary 33

In recent years, Long-Term Research Installation infrastructure funding by the Forest Science Program supported several activities: climate station monitoring by the CFS, maintenance of trails for field tours, updated signs, permanent photo points and replacement of cedar stakes with aluminum stakes for plot locations. FSP funding in 2008/09 supported 15-year measurement of planted and natural conifer regeneration, and understory vegetation. Long-term plots were measured one year before harvesting and 1, 3, 5, 10 and 15 years post-harvesting. The final report on this work was completed in April 2009.

The planted conifer study included four post-planting treatments: fertilization, vegetation control, vegetation control with fertilization, and no treatment. Five conifer species were planted at MASS: amabilis fir (Ba), western hemlock (Hw), western redcedar (Cw), yellow-cedar (Yc) and Douglas-fir (Fd). Only Hw and Ba received post-planting treatments. For natural regeneration and understory vegetation studies, tagged tree measurements and visual estimates of vegetation cover were done on 270 permanent plots.

Natural and planted conifer growth measurements over 15 years showed that: • There were no significant differences in tree survival among systems, although trends for planted Ba and Hw suggest lower survival in the shelterwood. Among the five species planted, Fd had lower survival (68%) than all other species (85%). The range in survival was similar for planted seedlings and advanced natural regeneration of Ba and Hw over the study period (71% to 96%). • Height and diameter growth of natural Hw was slower in the shelterwood compared to the more open clearcut, patch cut and green tree retention systems. Natural Ba growth did not differ significantly by silvicultural system. • For planted trees, the only differences in growth among systems were between the shelterwood (slower growth) and all other systems; however, significant differences were found only for some comparisons. • Total stem volume of natural Ba exceeded planted Ba for all systems; however, total stem volume of planted Hw exceeded natural Hw for all systems. There was no difference in seed source between planted Ba and Hw. Although genetics may play a role, the difference between Ba and Hw appear to be related to the ability of planted trees to establish root systems and acclimate to the site. • There was no significant relationship between age of advance regeneration and post- harvest height growth for either Ba or Hw. • Observations at MASS show that some Ba trees, and possibly Hw, have experienced reduced height increment or “growth check” in recent years (see report by Koppenaal et al. 2009).

Findings suggest the following silvicultural implications: • Small patch-cuts and low levels of dispersed green tree retention do not show significant adverse impacts on early survival and growth of montane conifers. • Results to 15 years suggest that planted Hw would be the most productive for regenerating shelterwoods or other shaded environments followed by natural Ba. • Modification of shelterwood design (e.g., strips, groups or lower uniform densities) may be required to allow sufficient light penetration for the release of advance regeneration.

Although the site conditions at MASS are representative of many montane sites in coastal BC, resource managers should be cautious about extrapolating the results to other variants or to drier sites within the CWHmm2. The scope of the MASS project is broad, covering a wide range of topics from economics to biodiversity with over 20 completed and active studies (website: www.cfs.nrcan.gc.ca/subsite/mass).

Western Forest Strategy – AM Summary 34

The Adaptive Management Cycle

This section of the report will be completed in Phase II of the project. It is important to document how the AM process has worked (or not) over the past decade. Often the feedback from the program is not as formal or linear as designed. Different elements of the strategy also have different time frames for completion of the AM cycle (i.e., Assess, Design, Implement, Monitor, Evaluate). Two elements of the forestry strategy completed one AM cycle: zoning and stand- level retention. The process and learning from this experience will be described in detail in the next phase. The early half of this “story” was told in Bunnell and Dunsworth (2009).

Some broad conclusions have emerged from our experience with variable retention:

• The landscape context determines what is necessary or appropriate for stand-level retention in relation to biodiversity conservation goals. • It is not practical to precisely mimic or emulate natural disturbance patterns. • Riparian networks are a governing factor for retention patterns. • Wind damage is a significant challenge for dispersing stand-level retention, making clearcutting with reserves the most viable option for some sites and landscapes. • Monitoring findings indicate that retention can achieve a lifeboating function for some organisms, with a positive correlation between patch size and species survival. • Retention has potential long-term benefits for enhancing structural diversity of second- growth forests. • Growth impacts on forest regeneration increase with greater dispersion of single trees or small groups. • Harvesting with retention can be done safely across a wide range of forest types and terrain using a variety of logging and silvicultural systems. • Costs are a significant challenge for applying variable retention on some sites in today’s marketplace. • The public visual preference for dispersed retention conflicts with ecological goals and operational needs that typically favour group retention.

Key advice for others considering adaptive management:

• AM requires a strong commitment from corporate leadership to both the approach and the resources to do it; • Operational staff must have sufficient “buy-in” to the approach (this will not happen immediately, so training and leadership are essential); • Partner with others. Seek outside advice and collaboration; no one can afford to go it alone; • Don’t try to measure or monitor everything; it is better to do a few things well than a lot of things poorly. [From a scientific advisory panel member.]

Western Forest Strategy – AM Summary 35

Adaptive Management Plan (2010-2014)

This section of the report will be completed in Phase II of the project. Recommendations will be made on: • Implementation monitoring; • Ecological representation; • Habitat structure monitoring; • Species monitoring; • Species accounting system and modelling; • Collaboration with others.

This plan will provide guidance for continuation of the Western Forest Strategy into the next decade.

Western Forest Strategy – AM Summary 36

References

Aubrey, K.B., C.B. Halpern and D.A. Maguire. 2004. Ecological effects of variable-retention harvests in the northwestern United States: the DEMO study. For. Snow Landsc. Res. 78(1-2): 119-137.

Bancroft, B. and K. Zielke. 2004. Implementation monitoring: five-year summary of variable retention on Weyerhaeuser BC Coastal Timberlands. Symmetree Consulting Group, West Vancouver, BC, 54 p.

Beese, W.J., B.G. Dunsworth, K. Zielke, and B. Bancroft. 2003. Maintaining attributes of old-growth forests in coastal B.C. through variable retention. For. Chron. 79(3): 570-578.

Beese, W.J., B.G Dunsworth and N.J. Smith. 2005a. Coast Forest Strategy implementation summary: 1999 – 2005. Weyerhaeuser, BC Coastal Timberlands, Unpublished report, 19 p.

Beese, W.J., B.G Dunsworth and N.J. Smith. 2005b. Variable retention adaptive management experiments: testing new approaches for managing British Columbia’s coastal forests. In: Balancing Ecosystems Values: Innovative Experiments for Sustainable Forestry. USDA For. Serv., PNW Res. Stn., Gen. Tech. Rept. PNW-635, pp. 55-64.

Bunnell, F.L., B.G. Dunsworth, D. Huggard and L.L. Kremsater. 2003. Learning to sustain biological diversity on Weyerhaeuser’s coastal tenure. Contract report to Weyerhaeuser, BC Coastal Timberlands. See: www.forestry.ubc.ca/conservation/forest_strategy

Bunnell, F.L. and B.G. Dunsworth. 2009. Forestry and biodiversity: learning how to sustain biodiversity in managed forests. UBC Press, Vancouver, BC, 341 p.

Campbell, W. and M. Preston. 2008. Breeding bird surveys on the southwest coast of British Columbia: 2000-2007 project summary. Final report to Western Forest Products, Westcam Consulting Services, Victoria, BC, 46 p.

Chan-McLeod, A. 2008. An experimental study of variable-retention harvest methods on forest birds. 2007-08 Annual Report to Western Forest Products, UBC Faculty of Forestry, Vancouver, BC, 25 p.

Clayoquot Scientific Panel. 1995. Scientific Panel for Sustainable Forest Practices in Clayoquot Sound, Report 5, Sustainable ecosystem management in Clayoquot Sound: planning and practices, Cortex Consultants Inc., Victoria, B.C., 296 p.

Coast Information Team. 2004. Integrated Land Management Branch, Province of British Columbia, Website: http://ilmbwww.gov.bc.ca/citbc/index.html

Franklin, J.F., D.R. Berg, D.A.Thornburgh and J.C. Tappeiner, J.C. 1997. Alternative silvicultural approaches to timber harvesting: variable retention harvest systems. In: Kohn, K.A.; and Franklin, J.F. eds. Creating a Forestry for the 21st Century: The Science of Ecosystem Management, Island Press, Washington, D.C., pp. 111-139.

Holling, C.S. 1978. Adaptive environmental assessment and management. John Wiley and Sons, New York, 377 p.

Huggard, D. 2006. Habitat monitoring 1999 to 2006—summary and data report. Unpublished report prepared for Western Forest Products, Adaptive Management Program, Campbell River, BC, 153 p.

Huggard, D.J. and F.L. Bunnell. 2007. Stand-level retention and forest birds: a synthesis of studies. Univ. of BC, Centre for Applied Conservation Research, Forest Sciences Centre, Vancouver, BC, www.forestbiodiversityinbc.ca

Western Forest Strategy – AM Summary 37

Iles, K. and N.J. Smith. 2006. A new type of sample plot that is particularly useful for sampling small clusters of objects. For. Sci. 52:148-154.

Lindenmayer, D.B. and J.F. Franklin. 2002. Conserving Forest Biodiversity: A Comprehensive Multiscaled Approach. Island Press, Washington, 349 pp.

Kremsater, L.L., F.L. Bunnell, F.L., D. Huggard and B.G. Dunsworth. 2003. Indicators to assess biological diversity: Weyerhaeuser’s coastal British Columbia forest project. For. Chron. 79(3):590–601.

Marmorek, D.R., D.C.E. Robinson, C. Murray and L. Greig. 2006. Enabling Adaptive Forest Management – Final Report. Prepared for the National Commission on Science for Sustainable Forestry by ESSA Technologies Ltd., Vancouver, B.C. 93 pp. Online at: http://ncseonline.org/CMS400Example/uploadedFiles/NCSSF/

Manning, E.T. 2008. 2007. forest songbird monitoring of WFP TFL 37, Woss, BC. Report by Manning, Cooper and Associates for Western Forest Products Inc,, Englewood Division, BC Forest Investment Account No. 6654003, 22 p.

McLennan, M.H. and P.E. Hennon. 2005. Maintaining old-growth features in forests used for wood production in southeast Alaska. In: Balancing Ecosystems Values: Innovative Experiments for Sustainable Forestry. USDA For. Serv., PNW Res. Stn., Gen. Tech. Rept. PNW-635, pp. 127-133.

Mitchell, A.K., R. Koppenaal, G. Goodmanson, R. Benton and T. Bown. 2007. Regenerating montane conifers with variable retention systems in a coastal British Columbia forest: 10-year results. For. Ecol. Manage. 246: 240-250.

Mitchell, A.K., A. Vyse, D.J. Huggard and W.J. Beese. 2004. Long-term silviculture experiments contribute to science-based forest management in British Columbia's public forests. For. Snow Landsc. Res. 78(1-2):139- 150.

Mitchell, S.J. and W.J. Beese. 2002. The retention system: reconciling variable retention with the principles of silvicultural systems. For. Chron. 78(3): 397-403.

Outerbridge, R.A. and J.A. Trofymow. 2004. Diversity of ectomycorrhizae on experimentally planted Douglas- fir seedlings in variable retention forestry sites on southern Vancouver Island. Can. J. Botany 82(11):1671-1681.

Outerbridge, R.A. and J.A. Trofymow. 2008. How do different levels of green tree retention affect the survival of ectomycorrhizae on the southern coast of British Columbia. 2007-08 Progress report prepared for the BC Forest Science Program, Canadian Forest Service, Victoria, BC, 46 p.

Ovaska, K. and L. Sopuck. 2008. Terrestrial gastropods as focal species for monitoring ecological effects of variable-retention logging practices. Final report to Western Forest Products and the BC Forest Science Program, Biolinx Environmental Research Ltd., Sidney, BC, 48 p.

Pearsall, I.A. 2008. Study to assess the efficacy of ground beetles (Coleoptera: Carabidae) as ecological indicators in two variable-retention experimental sites: a temporal comparison. Progress report prepared for the BC Forest Science Program, Pearsall Ecological Consulting, Nanaimo, BC, 166 p.

Preston, M.I. and A.S. Harestad. 2007. Community and species responses by birds to group retention in a coastal temperate forest on Vancouver Island, British Columbia. For. Ecol. Manage. 243: 156–167.

Preston, M.I. and R.W. Campbell. 2009. Monitoring birds for sustainable forest management: species- habitat associations and population trends for the southwest coast of British Columbia, 2000-2008, Report by Westcam Consulting Services, Victoria, BC. to Western Forest Products, Forest Investment Account Project No. 6743002, 86 p.

Western Forest Strategy – AM Summary 38

Rosenvald, R. and A. Lõhmus. 2008. For what, when and where is green-tree retention better than clearcutting? A review of the biodiversity aspects. For. Ecol. Manage. 255: 1-15.

Rollerson, T., C.M. Peters and W.J. Beese. 2008. Variable Retention Windthrow Monitoring Project: 2001-2008. BC Forest Investment Account Proj. No. 6458008, Golder Associates Ltd. and Western Forest Products, Campbell River, BC, 48 p.

Sadler, K. 2004. Vegetation monitoring in coastal Douglas-fir forests of Vancouver Island: influence of age class and edge proximity on vascular plant and bryophyte distributions. Report prepared for Weyerhaeuser, Nanaimo, BC.

Smith, N.J. 2009a. Effects of low levels of dispersed retention on the growth and survival for Douglas-fir, Draft manuscript, 22 p.

Smith, N.J. 2009b. Edge and experimental plots, Final report to Western Forest Products, BC Forest Investment Account, Proj. No. 6755022, 16 p.

Stanger, N. 2004. Edge and age effects on epiphytes in a lowland Douglas-fir forest. Report prepared for Weyerhaeuser, Nanaimo, BC.

Stankey, G.H., R.N. Clark and B.T. Bormann. 2005. Adaptive Management of Natural Resources: Theory, Concepts, and Management Institutions. Gen. Tech. Rep. PNW-GTR-654. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station. 73 pp.

VILUP. 2000. Vancouver Island Summary Land Use Plan. Province of British Columbia, 204p. Online at: http://ilmbwww.gov.bc.ca/slrp/lrmp/nanaimo/vancouver_island/plan/summary_lup/toc.htm

Wind, E. 2008. Pre- and post-harvest amphibian and small wetland study: 4-year results. Progress report to Western Forest Products and Island Timberlands, E. Wind Consulting, Nanaimo, BC, 28 p.

Western Forest Strategy – AM Summary 39

Appendix 1

Western Forest Strategy Final Implementation Version July 2007

Western Forest Strategy – AM Summary 40

Western Forest Strategy Final Implementation Version Approved 24 July 2007 W.J. Beese, Forest Ecologist, WFP Corporate Forestry

Background

As a result of recent acquisitions, Western Forest Products Inc. (WFP) manages nearly 1.5 million hectares of productive forest land on the BC coast with an annual allowable cut of over 7 million cubic metres of timber. Most of this area is public (Crown) land managed under Tree Farm Licenses (TFL) granted by the BC government. The three former companies—WFP, Canfor and Cascadia—had different approaches to forest management, particularly in the application of variable retention harvesting. The goal of the Western Forest Strategy is to bring operations from the legacy companies together under a common strategy that continues our commitment to Sustainable Forest Management (SFM), including the goal to conserve biodiversity on managed forest lands. This will be accomplished in a way that supports a safe, profitable and socially responsible business. The strategy builds upon the experience and success of the legacy companies while recognizing the value of a diversity of approaches in different geographical areas and forest types. This strategy was approved by senior management in July 2007.

Objectives

• To implement a unified company strategy to help meet our Forest Stewardship “Future State”: WFP maintains standards and practices to ensure that we are recognized as effective stewards of the resources and environment in which we operate. We achieve this in part by conserving biological diversity on Western Forest Products’ tenures. • To implement a policy on the use of variable retention silviculture and landscape zoning that is safe, profitable and socially responsible. • To manage harvesting costs in support of the company’s margin-focused business strategy while maintaining and improving marketplace acceptance of our forest management practices. • To conduct a monitoring program that verifies the implementation and effectiveness of the strategy and provides feedback for improvement.

Assumptions

The strategy was created with certain assumptions regarding safety, and the biological, social and economic elements of SFM:

Safety • At Western Forest Products, health and safety are integrated as a core business and personal value forming the framework in which we operate. Safe work is an expectation and pre-requisite for continuing operations. • Experience among legacy companies in recent years has shown that trees can be left during logging operations in a safe manner. • All guidelines for stand-level retention include the essential caveat that all employees have a right to a safe workplace and also have a responsibility to work safely. All plans and practices will have the flexibility to accomplish the job safely.

Western Forest Strategy – AM Summary 41

Biological • Coastal BC has a diversity of forest ecosystems and species; therefore, forest management practices must vary in response to that diversity. No single harvesting or silvicultural system is appropriate everywhere. Clearcut, seed tree, retention, shelterwood and selection systems are all ecologically appropriate in the right context. • Forest ecosystems and species have evolved in response to changes in climate and different natural disturbances (e.g. fire, wind, landslides, insects and disease) at various scales. While it is important to use scientific knowledge of historical development and habitat as a guide, we need not “mimic” natural disturbance to sustain productive and diverse forest ecosystems. We recognize the resilience of ecosystems and the multiple pathways and patterns that can occur within the limits of ecosystem processes. We also recognize that future conditions may differ from the past due to factors such as climate change. • Both stand-level retention and landscape-level reserves are necessary for conserving biodiversity across the landscapes with which WFP tenures are associated. Neither approach alone is likely to be as effective or efficient. Economic • A successful forest management strategy must maintain a profitable business for the sustained benefits to employees, shareholders and the public. The WFP strategy will improve profitability over the status quo of the three successor companies. • Variable retention (VR) is intentionally diverse in its approach; consequently, net costs compared to clearcutting (or clearcutting with reserves) range from cost savings to significant increases in costs. There are also opportunities to improve margin without violating silvicultural principles. While there is some information available on incremental VR costs, relative cost increases are not well documented. • There are economic benefits in the marketplace from publicly acceptable forest practices, including certification. Benefits accrue from maintaining or increasing market share (i.e., keeping existing customers and attracting new ones), and reduced costs for countering negative market campaigns or addressing public complaints. While these benefits are difficult to quantify, some examples from legacy company experience include: improved market share for redcedar products with Home Depot; Keurhout approval in the Netherlands, providing continued access to European markets; and reduced ENGO pressure from market campaigns. To date, there does not appear to be a price advantage directly attributable to SFM certification or environmental policies. Social • A strategy for conservation of biological diversity supports the company’s SFM certification. Certification is a cornerstone of public acceptance of our management practices, providing independent verification of sustainability and continuous improvement. It is also a component of the Stewardship goal in WFP’s Future State. • The lower profile on clearcutting in old growth in recent years is, in part, the result of the significant increase in the use of VR by companies in coastal BC and other jurisdictions. To avoid a public backlash on this issue, variable retention will continue part of the biodiversity strategy on WFP’s tenure. • Public support for our strategy can be enhanced by having endorsement of its biological validity from independent scientists.

This document presents a strategy for variable retention and zoning. It will form the basis of guidelines for operations. Further details on implementation will be developed by a working group with representatives from forest operations. A monitoring and adaptive management program to support this strategy, building upon the programs of the legacy companies, will be completed in 2007.

Western Forest Strategy – AM Summary 42

Terminology

The term variable retention (VR) is used to describe an overall approach to harvesting and silvicultural systems that retains trees and associated habitat for purposes other than timber management and traditional silviculture goals. Variable retention can be implemented with a wide range of harvesting systems, and can utilize traditional silvicultural systems, such as shelterwood or selection, to meet forest regeneration objectives. As the name implies, various levels of retention can be used with different types, amounts and spatial patterns of structure. Retention can be dispersed throughout a cutblock (as individual trees or small clumps) or aggregated in larger groups and patches, depending upon the objectives. There is such a wide range of possibilities within the VR concept that it is not a “one size fits all” approach.

The term retention system refers to a specific silvicultural system designed to meet the goals of variable retention. It was originally defined in the BC Operational Planning Regulations (March 1999) and has 3 requirements: 1) retention of trees distributed across the cutblock; 2) trees are left for the long term (at least one rotation); 3) distribution of leave trees achieves >50% “forest influence”. The specific definition of the retention system is: “a silvicultural system that is designed to:

a. retain individual trees or groups of trees to maintain structural diversity over the area of the cutblock for at least one rotation, and b. leave more than half the total area of the cutblock within one tree height from the base of a tree or group of trees, whether or not the tree or group of trees is inside the cutblock.” The distribution of long-term retention over the area of the cutblock is open to interpretation, but the spatial requirement in “b” for “forest influence” provides the minimum standard for distribution. The retention system is no longer officially defined in BC legislation; however the BC Forest Planning and Practices Regulation (Div.5, 64(4)) exempts harvesting that maintains >50% forest influence and meets other spatial requirements from maximum cutblock size restrictions. The retention system is considered a “partial cutting” approach and is categorized as an “even-aged” system despite the resulting uneven-aged forest because the cut areas are regenerated and managed much like other even-aged systems. (For further description see Mitchell and Beese (2002) or the on-line Introduction to Silvicultural systems at: http://www.for.gov.bc.ca/hfp/training/00014/chap2frt.htm)

The retention system normally uses a one-pass harvesting approach, but may also be prescribed with several harvesting entries. The three main variants of the retention system are: group, dispersed, and mixed. For safety, economic and ecological reasons, group retention is often preferred; however, all three variants have advantages for specific objectives.

Clearcut with reserves is a modification of traditional clearcutting where trees are retained within or adjacent to the cutblock. This approach differs from the retention system in two ways: 1) there is no spatial distribution requirement for the reserves; and 2) the reserves need not be left for the long term. In practice, clearcut with reserves often does leave long-term retention; therefore, it can be an effective tool within a landscape-level biodiversity strategy. Where at least some long-term reserves are distributed within the cutblock, the approach can be quite similar to the retention system in both appearance and function.

This strategy allows for a full range of silvicultural systems to be used for maintaining diverse forest conditions on the landscape, including both clearcut with reserves and the retention system. In this strategy, the term variable retention (VR) refers collectively to the retention system, shelterwood with reserves, and selection with reserves.

Western Forest Strategy – AM Summary 43

Pre-merger status of stand-level retention

Pre-merger use of variable retention is summarized in Appendix 1 by Region and Operation. The approach of the legacy companies to VR implementation is briefly summarized as follows:

• Cascadia Forest Products (formerly MacMillan Bloedel / Weyerhaeuser) VR was phased-in over a 5-year period (1999-2003) throughout all coastal operations under the “Coast Forest Strategy” (Forest Project) approved in June 1998. Over 90% of the area harvested since 2003 has been done using the retention system (99% in 2005). Stand-level targets for retention were set using a zoning scheme (Timber, Habitat, Old Growth), which specified the management emphasis for different landscapes. Average retention was 21% with half deemed to be incremental to government regulations. A monitoring and adaptive management program was developed with advice from UBC biologists and an independent scientific panel.

• Canadian Forest Products A retention strategy was developed for TFL37 as part of the “Forestry Principles” approved in June 1999 for coastal operations. Implementation in the field began in January 2001. Stand-level targets for retention were based on the Vancouver Island LUP zones (General, Enhanced, and Special Management), biogeoclimatic units and natural disturbance regimes (fire, gap). Over 90% of the harvested area included patch or single tree retention in 2005. At least 50% likely qualified as the retention system. (Note: 92% met the forest influence goal; however, only 20% was reported as retention or selection systems in GENUS). Stand-level Wildlife Tree Patch (WTP) reserves averaged 15%; however, “internal” patch retention was reported separately as 11%, but only a portion was designated as WTP. Assuming half was incremental to WTPs, total stand-level retention was about 20%.

• Western Forest Products Stand-level retention occurred throughout WFP coastal tenures predominantly through the clearcut with reserves approach. Wildlife tree patch requirements were established through higher-level plans and government guidelines. An average of 7% of the harvest area was done using the retention system across tenures, predominantly in TFL6 and TFL25. Average stand- level retention across all TFLs was 19%.

In addition to the company strategies summarized above, operations on the Central Coast of the BC mainland are part of the Coast Forest Conservation Initiative (CFCI). Through negotiation and collaboration with the BC government, First Nations, forest companies, communities and environmental organizations, a land-use decision was announced in February 2006 for a 6.4 million hectare region. Under this agreement, Ecosystem-Based Management (EBM) will be phased in by 2009 under guidelines developed by a scientific panel.

Western Forest Strategy – AM Summary 44

Strategy for Variable Retention

Key elements

The unified strategy for Western Forest Products’ approach to stand-level retention includes the following elements:

1. Maintain long-term retention on cutblocks using a combination of silvicultural systems, based on safety, ecological and operational factors; 2. Implement guidelines and targets for the amount of the retention system used in each WFP Timberlands Region and Operation based on zoning, ecological units, site and stand conditions, economics and harvesting systems; 3. Encourage a diversity of approaches throughout the company but maintain consistent standards; 4. Support the company’s margin-focused business strategy. Achieve overall cost reductions by: lowering the percentage of the retention system used on Cascadia areas but increasing it on legacy WFP areas, lowering the average cost of VR by applying clearcut with reserves on the most costly sites for the retention system, improving the cost-efficiency of VR implementation, and seeking opportunities for greater margin.

This strategy applies on Crown land and private land within TFLs outside of the Central and North Coast Ecosystem Based Management (EBM) planning area. Stand-level practices for the EBM area are defined by government planning processes. Private lands outside of TFLs are not included in this strategy.

Proposed approach

Briefly, the Resource Management Zones of the Vancouver Island Land Use Plan (VILUP) were used as the first level of stratification for guiding stand-level retention practices. Special Management Zones (SMZ) are mandated by government to use the retention system or clearcuts up to 5 hectares. General Management Zones (GMZ) use intermediate levels of retention and greater use of the retention silvicultural system, and Enhanced Management Zones (EMZ) have the lowest retention requirements and highest timber expectations. Within each RMZ, guidelines vary by ecological unit with lower targets for the retention system in windthrow prone areas. This strategy results in considerable variation in the use of the retention system (from 30% to 100% across strata) and variation in the minimum stand-level retention (10% to 20%) across WFP tenure. More details of the approach follow.

The provincial Ecoregion classification subdivides the BC coast into areas of similar physiographic attributes and climate. Ecosections further subdivide an Ecoregion based on sub-regional variation in these properties. This classification is a useful framework for dividing WFP’s tenure into broad ecological units that vary in climate, topography and natural disturbance regimes. Most of WFP’s operations fall within one or two Ecosections (Figure 1). Although there are two Ecoregions subdividing QCI/Haida Gwaii, most of the company tenure falls within the Skidegate Plateau Ecosection. For simplicity, the WFP Queen Charlotte Islands forest operation is represented by the abbreviation QC in tables.

Table 1 presents guidelines for use of the retention system by VILUP zone and Ecosection. WFP Forest Operations within each Ecosystem are also listed. The retention system target (%RS) will be applied on an area basis (i.e., hectares logged) rather than a volume basis. Shelterwood and selection systems with long-term reserves also count towards the targets.

In Enhanced Management Zones (where the emphasis is on timber production) the retention system will be used for 30% of the harvested area in the high wind environment of the northwest coast of Vancouver Island (Nahwitti Lowlands and the northern portion of the Windward Island Mountains Ecosection), and 50% of the harvested area in other Ecosections. The minimum long-term stand-level retention target is 10%.

Western Forest Strategy – AM Summary 45

Figure 1. WFP tenure showing Ecosection boundaries. NL=Nahwitti Lowlands, NIM=Northern Island Mountains, LIM=Leeward Island Mountains, WIM=Windward Island Mountains, SPR=Southern Pacific Ranges, QC=Queen Charlotte (includes QC Lowlands, Skidegate Plateau, and WQC Ranges Ecosections). LIM and WIM units are divided (unofficially) into North and South at Strathcona Park.

In General Management Zones (where the emphasis is on integrated resource management) the retention system will be used for 40% of the harvested area in the high wind environment of the northwest coast of Vancouver Island (Nahwitti Lowlands and the northern portion of the Windward Island Mountains Ecosection), and 60% of the harvested area in other Ecosections. The minimum long-term stand-level retention target is 15%.

In Special Management Zones (where specific environmental, recreational and cultural/heritage values have been identified) the VILUP Higher Level Plan Order specifies: “applying a variety of silvicultural systems, patch sizes and patch shapes across the zone, subject to a maximum cutblock size of 5 ha if clearcut, clearcut with reserves or seed tree silvicultural systems are applied, and 40 ha if shelterwood, selection or retention silvicultural systems are applied.” Cutblocks larger than these specifications may be approved for salvage operations and, wherever possible, “the cutblock incorporates structural characteristics of natural disturbances.” A minimum of 20% long-term stand-level retention is recommended for SMZs in the Western Forest Strategy based on both social and biological criteria. Studies of visual impacts of VR at UBC suggest that the public generally does not recognize group retention as different from clearcutting until retention levels are at least 20%. An analysis of studies for 69 species of birds by Huggard and Bunnell (2007) concluded that the local abundance of many less sensitive bird species decreases substantially below 15-20% retention. Average stand-level retention of

Western Forest Strategy – AM Summary 46

Table 1. Proposed guide to retention by Res. Mgt. Zone and ecological units

VILUP Special Management Zone (SMZ) Higher Level Plan Order specifies a maximum cutblock size of 5ha for clearcut, clearcut with reserves or seed tree systems and a 40ha maximum size for retention, shelterwood or selection systems. WFP SMZ Minimum long-term stand-level retention: 20%

VILUP WFP % RS Zone Ecosection Operations Target GMZ NL Holberg, Jeune, McNeill 40 WIM-N Gold R.,Zeballos, Nootka 40 * NIM McNeill, Englewood, Mid-Isl., Gold R. 60 WIM-S Pt.Alberni, Jordan R. 60 * LIM Mid-Island, Pt.Alberni 60 * SPR Stillwater 60 QC Queen Charlotte 60 Minimum long-term stand-level retention: 15% (20% in dm, xm, mm1) * 70% RS for CWHdm, xm, mm1 Rationale: Intermediate %RS in GMZ; < targets in windward NVI areas; > % retention in dry variants (R&B listed).

VILUP WFP % RS Zone Ecosection Operations Target EMZ NL Holberg, Jeune, McNeill 30 WIM-N Gold R.,Zeballos, Nootka 30 * NIM McNeill, Englewood, Mid-Isl., Gold R. 50 WIM-S Pt.Alberni, Jordan R. 50 * LIM Mid-Island, Pt.Alberni 50 * SPR Stillwater 50 QC Queen Charlotte 50 Minimum long-term stand-level retention: 10% (15% in dm, xm, mm1) * 60% RS for CWHdm, xm, mm1 Rationale: Lower %RS in EMZ; < targets in windward NVI areas; > % retention in dry variants (R&B listed). % RS = Percent of harvested area using the Retention System or Shelterwood/Selection w/ reserves. Note: The N. & Central Coast EBM area is not covered by these guidelines. the three legacy companies was about 20%, so this target should not be difficult to achieve. WFP will implement the SMZ provisions initially using the interim NVIR SMZ Retention Strategy and Guidelines (developed for SMZs 2 and 4 in TFL 6), the TFL37 Forestry Principles strategy, or other local guidelines where applicable, until company-wide guidelines are developed.

Appendix 2 presents a summary of the total area and proportion of VILUP zones by WFP forest operation, average retention system targets by zone, and the estimated overall retention system goal for each operation and region. For operations on Haida Gwaii (QCI) and the lower coastal mainland (Stillwater), not covered by VILUP, the legacy Cascadia zones were used (i.e. Timber=EMZ, Habitat=GMZ, Old Growth and Recreation=SMZ). This approach will have to be confirmed with these operations in case adjustments of zoning are warranted. The company tenure within the North and Central Coast under EBM planning was not included in these guidelines because requirements will be set by government LUP agreements (currently at 15% stand-level retention and 30% to 70% of natural old forest representation at the landscape level). Note that Appendix 2 uses the total hectares in each zone. A comparison of total hectares to net productive hectares (THLB) showed little difference in the resulting percentages.

Western Forest Strategy – AM Summary 47

The overall approach of emphasizing the retention system in five of the seven Ecosections (or subdivisions) and emphasizing clearcut with reserves in the two Ecosections on the northwest coast of Vancouver Island results in a range of targets by ecological unit and management emphasis from 33% to 61% retention system by operation. Regions vary from 44% to 58%, with an overall average for the company of 51% of the total harvest area utilizing the retention system.

To maintain flexibility, given the need to adjust harvesting schedules for weather, markets and other conditions, the %RS goals will be applied over a 5-year period once phase-in of the strategy is completed. These targets may be refined by biogeoclimatic unit for some landscape units after a spatial analysis of representation of ecosystem types in Old Growth Management Areas (OGMAs) and other landscape-level reserves. Stand-level retention may be used to compensate for deficiencies in landscape-level representation of specific ecosystems. Extensive logging history and poor representation in reserves in leeward, drier biogeoclimatic variants (CWHdm, xm, mm1) is the rationale for setting targets that are higher for both the retention system and average percent retention for cutblocks in these variants. The overall target for the Zone/Ecosection, however, remains the same. Specific guidelines for implementation of zoning and stand-level retention will be developed by the VR Working Group during phase-in of this strategy. It is important to note that this is a forward-looking strategy. Targets for the amount of the retention system used by zone apply to current and future harvesting plans. Operations will not return to previously completed retention blocks and remove reserves, except as allowed under salvage guidelines developed by the VRWG.

Phase-in

The guiding principle with phase-in is a desire to avoid changes to cutblocks already planned, unless there are financial incentives to do so. Targets for the retention system will be phased-in from 2008-2010 under the following schedule:

YEAR TARGET (% OF HARVESTED AREA) 2008 At discretion of each Region/Forest Operation 2009 Half way towards targets 2010 Full implementation

For example, a legacy WFP operation with a target of 30% retention system (RS) in an Enhanced zone would have a target of 15% RS harvesting for 2009 and 30% RS for 2010. The 2008 target could be set between 0 and 15% depending upon the status of 2008 planning. Conversely, a legacy Cascadia operation currently at 100% RS with a target of 60% RS in a General zone would have a target of 80% RS for 2009 and 60% RS for 2010. The estimated change for legacy Canfor-Englewood is only 6%, so phase-in is less of an issue than possible adjustments to stand-level guidelines for leave trees, which will have to be determined with local input. As part of phase-in, we will have to examine existing commitments in Forest Stewardship Plans and SFM Certification plans to ensure consistency or amend these documents.

Biological Rationale

The shift from almost exclusive use of the retention system in legacy Cascadia operations to a mixture including clearcut with reserves is supported by monitoring activities. Results from several projects show: • Larger retention patches or reserves are less vulnerable to windthrow than smaller patches. • Few tall snags remain in typical (0.25 ha) retention groups because they are removed for safety reasons; retaining larger patches using “clearcut with reserves” will allow more safe snags to be retained. • Some organisms that are not very mobile (ground beetles, gastropods) are able to survive in retention patches and do not show significant benefit from single leave-trees until retention levels are high. A positive relationship was found between patch size and abundance (i.e., larger patches are better). Plants such as epiphytic lichens and bryophytes are likely to show similar relationships. However, part of the function of the retention system is to distribute structure from the original stand so that species depending on old-forest attributes will re-colonize the second-growth stand more readily.

Western Forest Strategy – AM Summary 48

• Bird studies also support the benefits of patches over single tree retention; some species show a preference for larger patches, but habitat features of the patch are also very important.

A mixture of group retention and clearcut with reserves will achieve a range of group sizes and provide for more large patches than current applications of the retention system alone. If large patches are generally better for more species than small patches, why not simply do all harvesting with clearcut with reserves? Biological arguments for variable retention include:

1. To achieve conservation of biological diversity, the basic theoretical premise is that species are adapted to historic local conditions including natural disturbances. In coastal BC, windthrow, insects, disease, infrequent fire and landslides create forests with an abundance of dispersed residual structure (e.g., live and dead standing trees) from the pre-disturbance stand. Large, even-aged patches do occur but are the exception rather than the norm. Therefore, we are more likely to conserve the range of species on our tenure if we apply harvesting practices that leave some residual trees spread throughout a cutblock. 2. Dispersal of residual trees facilitates movement and dispersal of various species. Examples include some canopy lichens and bryophytes (mosses and liverworts) that are unique to old- forests and may not spread easily. 3. Mycorrhizal fungi (important for conifer regeneration) are more abundant near retained single trees and forest edges. Rates of spread and re-colonization from older forests to second-growth are poorly understood but would be enhanced with retention. 4. For some species of birds, 15-20% retention distributed over a cutblock can maintain a similar abundance of their population to the uncut forest condition. For these species, we can leave fewer trees using the retention system than using clearcut with reserves to achieve the same habitat benefit. 5. Territorial species that require certain old forest attributes (such as large snags) will benefit from having these attributes distributed throughout second-growth stands rather than concentrated in a smaller area. The same level of retention may support more nesting pairs in an area if suitably distributed than if located in a single patch. 6. Finally, with many species on our tenure, and poor knowledge of the habitat requirements for most of them, we consider the advice of Dr. Fred Bunnell to be a practical approach for maintaining biodiversity: “don’t do the same thing everywhere.”

Although the goal is to have long-term retention associated with nearly every cutblock, either within or adjacent to each block, flexibility is needed for circumstances where stand-level retention is neither possible nor desirable. Exceptions to guidelines for long-term retention include: • Catastrophic events. It may be impossible or ill-advised to meet retention system requirements in cutblocks during salvage operations for windthrow, wildfire, insect or disease outbreaks because of safety and economic constraints or low likelihood of attaining ecological goals. • Hardwood management. For intensive plantations of poplar or alder that represent a very small portion of the entire landbase, the approach is clearly tree farming. In such cases, retention would compromise efficient and economically viable hardwood management. • Special site conditions: There may be situations where exceptions to the ‘general rule’ are the best option. Examples: a small cutblock with no internal or edge retention where large adjacent reserves already exist; an exposed ridge where soil, forest type and local wind conditions indicate a low potential for windfirm reserves. • Private land. The company has only a small amount of private land. At this time, the guidelines for use of the retention system do not apply outside of TFLs and FLs. While many of the ecological goals of retention are relevant to forest health and productivity, the impact on timber growth and the value of the land as a company asset are important factors in the balance of objectives. For private land sales, the goals for long-term retention no longer apply.

The WFP strategy will result in a wide range of stand conditions, snag retention and structural distribution over the landscape and allow more flexibility for dealing with windthrow and forest health concerns--while maintaining the basic principles of the variable retention approach.

Western Forest Strategy – AM Summary 49

Economic Rationale

Experience with implementation over the past 9 years has identified the factors leading to both cost increases and cost savings associated with variable retention. A partial list of these includes:

Factors associated with increased costs for variable retention

Planning and layout • Increased field layout, surveying, and mapping • Increased number of blocks to layout for the same volume vs. clearcuts (only where there is retention incremental to FRPA requirements) • More complex Silviculture Prescriptions

Falling • More time spent opening up faces • More edge to work against and to treat for hazard trees • Increased jacking to control direction of falling timber

Yarding Hoe Chucking • Yarding around retention patches can increase the overall yarding distance Cable • Extra rigging time for more frequent yarder moves • Frequent machine moves to work angles around retention patches • Loss of backspar trail opportunities Helicopter • Long line has to be used along timber edges, reducing productivity • Potential for longer flights to avoid retention

Roads • Less volume harvested from the same in-block road system required for a clearcut • Occasional need for extra within-block road to facilitate yarding

Factors associated with cost savings for variable retention

Planning and layout • No size limit for cutblocks done to Cascadia retention system standards under FRPA; this removes adjacency constraints on timber development, allows concentration of equipment and reduces the extra roads required for cutblock dispersal under clearcutting; • Ability to extract higher average sales value from the volume harvested by locating retention patches on areas with lower economic value, yet high biodiversity value (i.e., increase margin); • Access to volume that would not be approved for clear-cutting under government regulations; • Potential to reinstate stumpage allowance for the retention system (need to lobby government).

Falling • Ability to lower the cutblock falling cost (average per m3) by leaving difficult areas in retention (e.g., patches with higher snag density; rock outcrops); this can also reduce safety hazard.

Western Forest Strategy – AM Summary 50

Yarding Hoe • Opportunities to leave retention in areas of difficult yarding (e.g., wet ground, small areas of steep slope); • Retaining individual trees that lower productivity (e.g. oversize, low-quality vets). Cable • Leaving retention in areas of poor deflection. Helicopter • Similar opportunities for leaving groups or individuals trees that would otherwise increase yarding costs.

In addition to the cost-savings opportunities (compared to clearcutting) listed above, experience with retention systems has identified a number of ways to lower the impact on costs.

Approaches that reduce the cost of variable retention

Falling • Emphasis on group retention over dispersed; • Reduce the percentage of systems with greater productivity impacts, such as uniform or strip shelterwoods, group selection, or confine to ground-based systems; • Reduced emphasis on “visual” criteria for less visually sensitive areas, while maintaining the biological criteria for retention design.

Yarding General • Factors listed for Falling also apply; • Limit the amount of retention that is within 1.5 tree lengths from roadsides; • Maintain retention levels close to the zone minimum where this is consistent with overall goals; • Optimize road layout within cutblocks. Hoe Chucking • Avoid creating “pinch points” for log flow. Cable • Limit corridor and corridor/carriage yarding to up-hill whenever possible Helicopter • Keep retention patches > 1.5 tree lengths from landing zones • Design cutblocks with flight paths in mind • Access developed areas that could not be logged if clearcut

The strategy is estimated to result in net cost-savings for the company when fully implemented, based on estimates for each operating area using the land use and ecological zoning approach described above. The strategy reduces by 14% the amount of the retention system that was practiced by the three legacy companies (Table 2). The North Island Region will experience a 10% increase in the use of the retention system but the other two regions will experience a decrease—the greatest in the Campbell River Region where most of the legacy Cascadia operations occur. Implementation guidelines will allow flexibility to adapt practices to site conditions and logging systems.

Reducing legacy Cascadia operations from virtually 100% retention system to under 60% will achieve significant cost savings through both 1) the savings in incremental costs on about 40% of the harvested volume, and 2) lowering the overall average cost of logging by converting the majority of the most costly sites (e.g., cable yarding) from the retention system to clearcut with reserves. Some of the cost-savings will be used to bring legacy WFP operations up to an average of about 40% retention system. The biological rationale for this is described below.

Western Forest Strategy – AM Summary 51

Table 2. Percent retention system by hectares logged.

Pre- New Region Merger WFP North Island 40 50 West Island 53 44 Campbell R. 99 58

Total 65 51

This document does not attempt to estimate potential timber supply impacts; however, it seems reasonable to assume no worse than a “no net change” because all three legacy companies were leaving similar amounts of stand-level retention. There are opportunities to reduce the amount of retention through consistent application of guidelines. In addition, the change from the Cascadia “Old Growth Zones” (where 66% of the area was set aside as old growth reserves) to VILUP Special Management Zones will result in positive impacts on timber supply and should more than compensate for any reductions to THLB in other legacy operations. Detailed analysis must be done to test these assumptions and make any modifications to TFL AAC determinations.

Social Rationale

The proposed strategy seeks to achieve public and marketplace support for our practices by demonstrating a commitment to conserve biological diversity on our tenure. We believe that the WFP strategy will garner public support by maintaining independent CSA SFM certification and: • Continuing to implement a variable retention approach to forest management; • Using the retention silvicultural system for the majority of the company’s harvesting, and maintaining consistent standards for its application throughout the company; • Using clearcut with reserves with long-term retention to meet biological goals where windthrow or high costs make other systems impractical; • Managing for visual aesthetics in areas of high visual concerns; • Demonstrating the validity of the strategy through monitoring and research, and changing as our knowledge grows (i.e., adaptive management).

We believe this strategy will meet marketplace expectations while improving safety, costs and productivity for areas where implementation of the retention system is most challenging (e.g., cable yarding, high windthrow landscapes).

Zoning

The basic concept behind zoning is to create different intensities of management in different landscapes under the assumption that a variety of conditions will support more effectively a variety of organisms. The other purpose is to recognize the ecological differences in the type and frequency of natural disturbances that have shaped the seral stage distribution and stand structural conditions in a landscape. Zoning is a useful framework for applying different regional or landscape-level management goals for the company’s tenure. Various government land use objectives in higher-level plans (e.g. the Vancouver Island Land Use Plan (VILUP); Biodiversity Emphasis Options (BEOs)) already set different targets by landscape unit. Cascadia and Canfor applied additional zoning schemes to set targets for stand-level retention and other practices. Some areas (e.g. the Central Coast EBM handbook) already have a set of guidelines that require a different approach than other areas of our tenure.

Western Forest Strategy – AM Summary 52

The Western Forest Strategy uses existing BC Government ecological and administrative zoning as a framework to apply guidelines for the type and amount of stand-level retention across the new WFP landbase. It is a modeled after the biodiversity conservation strategy developed by Canfor’s Englewood division that combined management intensity zoning (VILUP), biogeoclimatic units and dominant disturbance types to apply different stand-level retention guidelines to the landscape. Because of the broad scale covered by WFP operations, Ecosections were used instead of BEC for dividing our tenure into ecological units that vary in climate, topography and natural disturbance regimes.

Forestry planners in the legacy Cascadia operations may wonder why the new WFP strategy did not use the Timber, Habitat and Old Growth stewardship zones from the Coast Forest Strategy. First, the combination of VILUP zones and Ecosections provided a familiar framework that could be applied across all tenures on Vancouver Island. Second, the original CFS zones were no longer serving the original intent for the following reasons:

1. Changes in the landbase from company sales, government tenure removals and new protected areas rendered the former zoning allocations obsolete. Nearly all private lands removed to form Island Timberlands were in the Timber zone. Significant Old Growth zone areas were removed in the “20% take-back” from TFLs 39 & 44, as well as in the Iisaak tenure in Clayoquot. Some former TFL areas were designated protected areas (e.g., Koeye, Namu). The remaining area from former Weyerhaeuser tenures accounts for only about 60% of the new WFP combined operations (~850k/1400k ha).

2. Old growth zones were never implemented as planned. Operations did not use the multi-aged stand management approach envisioned for the Old Growth zones because the costs were too high under current conditions. Old Growth zones effectively became 100% reserves. Even before the company sales and tenure changes, the original target of 10% for Old Growth zones was never achieved (actual was 8%). The original concept of zones applied to large landscapes (e.g. watersheds) was changed to “mini-zoning” in several areas in order to meet local areas of concern.

3. Zone designations had a weak ecological basis. Other than an attempt to distribute each zone throughout the company tenure, zone locations had more to do with the level of existing development and land ownership than natural disturbance regime or ecological conditions in the landscape. Although the amount of remaining old growth forest was an overriding factor for designating Old Growth zones, retention standards for other zones were not related to biogeoclimatic units, NDTs, fire or wind disturbance history or other ecological criteria.

As noted above, for areas not covered by VILUP (Stillwater, QCI), the legacy Cascadia zone boundaries were used as an initial framework for applying the retention system guidelines (i.e. Timber=EMZ, Habitat=GMZ, Old Growth and Recreation=SMZ). This approach will have to be confirmed with these operations in case adjustments of zoning are warranted.

Monitoring and Adaptive Management

The key environmental objective of the Western Forest Strategy is to sustain biological richness and ecosystem productivity. The key economic objective is to obtain positive returns on investment in the company’s forest operations. Together these objectives are intended to sustain healthy, biologically diverse forests that ensure economic returns in the future. Because of uncertainty, an essential part of implementing the strategy is to use an adaptive management approach.

Adaptive management (AM) is a formal process for continually improving management practices by learning from the outcomes of operational programs. Monitoring is an essential component and is composed of two elements: performance and effectiveness. Performance monitoring measures whether or not you have met the stated targets and standards. Effectiveness monitoring measures whether or not you have achieved the desired outcomes.

Western Forest Strategy – AM Summary 53

An Adaptive Management Working Group (AMWG) was formed for legacy Cascadia tenures to develop a framework, methodology and pilot protocols for effectiveness monitoring. The group included members from the company, the Ministry of Forests, the Ministry of Environment, the Centre for Conservation Biology at UBC and private contractors. We envision this group continuing to advise the company’s AM program. The monitoring framework developed by the group outlines both extensive and intensive activities. The extensive or “passive” AM framework consists of monitoring structure and the presence or absence of species along with forest growth, health and windthrow in current and future cutblocks. The intensive or “active” AM framework consists of five designed comparisons replicated three times and focused on specific stand-level questions using an experimental approach. Treatments are operational scale (20-hectare units), so each area covers 80 to 100 hectares. Nine of fifteen sites were established, so one of the tasks for the WFP AMWG will be to examine the benefits of completing the remaining six experimental sites.

The task for this year will be to combine the existing elements of the monitoring and research activities of the legacy companies into an overall monitoring and adaptive management program. Combining these efforts should result in cost-savings over the expenditures by the separate companies. Funding by WFP and outside sources (e.g., the Forest Investment Account (LBIP, FSP), NSERC) will support this program and deliver a credible, scientific justification of our practices. The goal is to have an adaptive management and monitoring program for the new WFP landbase in place by 2008. This is an important element of our SFM Certification.

Linking monitoring back to management action is a fundamental component of an effective operational AM program. Key elements of the program include:

• An AM framework with criteria and indicators to guide the program; • VR and AM working groups to provide input on implementation and monitoring; • Implementation monitoring on a random sample of cutblocks; • Monitoring forest structural attributes for different approaches to variable retention and for benchmark sites; • Monitoring the impacts of VR on forest growth and yield, windthrow and small streams; • Monitoring forest birds at stand and landscape levels; • Completing pilot studies for a variety of species groups and developing indicator species for continued monitoring; • Establishing experimental areas for comparing the biological and economic impacts of variable retention; • Collaborated with government, universities, NGOs and other forest companies on research and monitoring initiatives.

Implementation

The following aspects must be addressed for effective implementation of the strategy:

1. Cost accounting. To measure progress towards cost-reduction goals we need a system for tracking or sampling costs. 2. Training. Efficient and safe implementation of VR in operations that have had little experience with the approach will require training. One cost-effective means of accomplishing this for field planners would be to organize job exchanges for a week or more between legacy company staff with different levels of experience with layout for the retention system. Training programs for other forest workers must be implemented. 3. Independent verification. Endorsement of our strategy from independent sources (e.g., academic advisors) will enhance our credibility. A few scientists could review the WFP strategy and provide feedback without conducting expensive workshops or panels. 4. Stakeholders and communities. The company’s Community Advisory Groups will need to be informed of the new strategy. These forums can also provide initial feedback on the level of public acceptance of our approach.

Western Forest Strategy – AM Summary 54

5. Customers and marketplace. Materials will have to be developed for company marketing staff so that they can answer questions about our forest management practices accurately and to ensure that we achieve maximum benefits from our strategy. 6. Guidelines and targets. We must balance simplicity with flexibility and site specific needs. The VR Working Group will need to examine and refine guidelines as needed based on biological, silvicultural and operational factors. Windthrow is an important factor for many operations; results of monitoring have already led to improved planning and layout to minimize wind damage. 7. Zoning. We will have to confirm the zoning assignments for areas not on Vancouver Island and develop a common GIS layer for zoning.

Guiding Principles

Finally, a mission statement and set of guiding principles for forest management by Western Forest Products would help unite our diverse operations and support the common vision in WFP’s Future State. Such a statement would clarify what we stand for and help guide our day-to-day actions and long-term objectives. This document would be best developed with input across the entire WFP Timberlands organization for effective buy-in. A draft of what this document might look like is given in Appendix 3.

Acknowledgements

This strategy is the result of considerable discussion among WFP employees and a few independent consultants. Many people deserve thanks for their review of previous drafts, suggestions, field trips and constructive debate. Special thanks to Wanda Kuzenko for creating the map in Fig. 1 and for GIS analysis, and to Paul Bavis for many helpful review comments.

Original Draft: September 18, 2006 Final Revision: June 28, 2007

Western Forest Strategy – AM Summary 55

Appendix 1. Estimate of % Retention System use in pre-merger and new WFP strategy (by volume).

Estimated New TotalPre-Merger Operations Western Strategy Volume RS CC/CCR Total % RS % RS Est. RS RS RS 2009 Region TFL Operation (m3/yr) (ha) ha ha (area) (vol.) (m3/yr) % (m3/yr) Target North VI 39 Pt. McNeill 350,000 711 2 713 100 100 349,018 57 199,500 79% 6 Pt. McNeill 295,000 53 523 576 9 9 27,144 33 97,350 17% 6 Holberg 645,000 53 551 604 9 9 56,598 45 290,250 23% 6 Jeune Landing 290,000 53 590 643 8 8 23,904 39 113,100 20% 37 Englewood 795,000 500 405 905 55 55 437,250 61 484,950 58% Sub-NVI 2,375,000 1,370 2,071 3,441 40 38 893,914 50 1,185,150 44% West Island 19 Gold River 496,000 0 712 712 0 0 0 42 208,320 21% 19 Zeballos 160,000 15 192 207 7 7 11,594 44 70,400 22% 19 Nootka Sound 444,000 0 577 577 0 0 0 38 168,720 19% 44 Port Alberni 852,000 1822 35 1857 98 98 835,942 58 494,160 78% 25 Jordan River 160,000 28 133 161 17 17 27,826 53 84,800 27% Sub-WI 2,112,000 1,865 1,649 3,514 53 41 875,362 49 1,026,400 45% Campbell River 39 Stillwater 405,000 677 0 677 100 100 405,000 58 234,900 79% 39 Mid-Island 992,000 2121 29 2150 99 99 978,620 56 555,520 78% 39 QCI 500,000 827 6 833 99 99 496,399 59 295,000 79% ** Sub-CR 1,897,000 3,625 35 3,660 99 99 1,880,018 57 1,085,420 78% TOTAL 6,384,000 6,860 3,755 10,615 65 57 3,649,294 52 3,296,970 55% By Pre-Merger Cascadia 3,099,000 6,158 72 6,230 99 99 3,064,978 57 1,779,080 78% Company Canfor 795,000 500 405 905 55 55 437,250 61 484,950 58% Western 2,490,000 202 3,278 3,480 6 6 147,066 41 1,032,940 21% Notes: % RS = Percent of harvested area using the Retention System or Shelterwood/Selection w/ reserves. This comparison is approximate only; actual RS% in the strategy is defined by VILUP RM Zones and ecological units, based on ha cut. Total Volume is based on the Timberlands Regional Structure summary (June 2006). Actual pre-merger volumes are not shown; volumes are estimates based on the % of total ha logged. Note that the pre-merger total % RS is 57% (by volume) rather than 65% (if calculated by area). applying the area % targets to the total volume for each operation may not reflect actual volumes. Cascadia Pre-merger stats based on 2005 Summary Canfor Pre-merger stats based on expert opinion of what likely met RS, even if not claimed on prescriptions. TFL37 (2005): 92% of blocks met 'forest influence' objective Western legacy Pre-merger stats based on 2004 Annual Reports for TFL6, TFL19 and TFL25 ** Mainland Coast=Stafford Lake (Blk 2), Swanson Bay (Blk 5); total volume 454,000 m3/yr; under EBM guidelines. Mainland Coast logged 64 ha/ 127 ha as VR (50%) in 2004; it was not included in the strategy or cost estimates. Pt.McNeill volume split between legacy WFP and Cascadia is estimated TFL6 RS split evenly among operations; logged area was 2006.6 ha; split based on 2109.8ha SPs completed

Western Forest Strategy – AM Summary 56

Appendix 2. Estimate of % retention system (area) by operation and VILUP zones. Guidelines Avg. % Total Hectares Zone Percent % Retention Sys. RS

OPERATION Enhanced General Special Total EMZ GMZ SMZ EMZ GMZ SMZ ALL Englewood 66,704 61,382 29,814 157,899 42 39 19 50% 60% 90% 61 Holberg 48,976 35 16,311 65,322 75 0 25 30% 40% 90% 45 Jeune Landing 53,367 8,123 2,962 64,452 83 13 5 35% 45% 90% 39 Port McNeill TFL6 50,933 17,928 0 68,861 74 26 0 30% 40% 90% 33 Port McNeill TFL39 15,089 33,354 0 48,444 31 69 0 50% 60% 90% 57 Gold River 55,348 49,851 5,793 110,992 50 45 5 35% 45% 90% 42 Zeballos 7,619 32,369 5,011 44,999 17 72 11 30% 40% 90% 44 Nootka 108,278 35,151 16,811 160,240 68 22 10 30% 40% 90% 38 Port Alberni 75,104 74,625 15,514 165,243 45 45 9 50% 60% 90% 58 Jordan River 24,269 4,173 1,241 29,683 82 14 4 50% 60% 90% 53 Mid-Island 107,106 33,224 14,476 154,806 69 21 9 50% 60% 90% 56 Stillwater 78,836 50,165 16,398 145,399 54 35 11 50% 60% 90% 58 Queen Charlotte 154,318 9,286 47,411 211,015 73 4 22 50% 60% 90% 59

Northern VI Region 235,069 120,823 49,086 404,978 58 30 12 50 West VI Region 270,619 196,170 44,369 511,158 53 38 9 44 Campbell River Region 340,260 92,675 78,285 511,220 67 18 15 58

Grand Total 845,948 409,667 171,741 1,427,356 59 29 12 51

Note: Operations outside VILUP used the legacy Cascadia zones; Timber=EMZ, Habitat=GMZ, Old Growth/Recn=SMZ. Area of private land and other misc. categories was excluded from this summary.

Western Forest Strategy – AM Summary 57

Appendix 3. Draft Forest Management Mission, Principles and Values.6

Western Forest Products Inc. Forest Stewardship Mission, Principles and Values

Our Mission

We, the employees of Western Forest Products, are committed to: • The safety and well-being of all employees and their communities; • Forest stewardship that sustains productive and diverse forest ecosystems; and • Producing quality products based on sound economic and environmental principles. Our Principles and Values The guiding principles and values for Western Forest Products’ forest management are: safety, stewardship, sustained economic and social benefits, responsibility and integrity. Safety means: Treating safety and occupational health as an uncompromised right and responsibility of everyone working for Western Forest Products. Forest Stewardship means: CONSERVATION—Conserving or enhancing forest productivity; air, soil and water resources; and biological diversity. OLD GROWTH—Sustaining old-growth attributes, natural features and wildlife habitat on managed forest landscapes through stand-level retention and landscape reserves. FOREST HEALTH—Maintaining long-term forest health, functions, vigor, genetics and timber quality through our harvesting and silvicultural practices. REGENERATION—Regenerating promptly with a diversity of ecologically suitable species over the landscape. SPECIES PROTECTION—Protecting the viability of native species through habitat management, and helping with recovery efforts for endangered species on our forest tenures. IMPROVEMENT—Continually improving our forest management activities and practices by monitoring results and incorporating new knowledge and experience. RECYCLING—Conserving natural resources through reducing, reusing and recycling materials. Sustained Economic and Social Benefits means: COMMUNITY—Being good neighbours in the communities in which we operate by actively seeking meaningful public input and involvement. SUSTAINABILITY—Providing a sustained timber harvest and maintaining future opportunities for products, jobs, cultural use, recreation and other benefits on the public lands under our care. BALANCE—Managing our forests to meet both shareholder and public expectations. INNOVATION—Supporting research and development to foster learning, innovation and continued improvement of forest management. Responsibility and Integrity means: ETHICS—Conducting ourselves with integrity and treating our employees, customers, colleagues, First Nations, communities, competitors, suppliers and shareholders with respect. COMPLIANCE—Meeting or exceeding all applicable laws, regulations and commitments. AUDITS—Using independent audits to evaluate and improve our forest management performance. TRAINING— Ensuring employees receive the training and resources needed to carry out their responsibilities competently and safely. COMMUNICATION—Sharing information and best practices among our forestry operations and improving communication with the public.

6 In support of the WFP “Future State December 31, 2009” document.

Western Forest Strategy – AM Summary 58

Appendix 2

Variable Retention Adaptive Management (VRAM) Experimental Comparisons

Group Retention: Variable % of area Dispersed Retention: Variable basal area retained; group sizes range from 0.2 to 0.5 ha retained; single trees or small groups up to 0.1 ha

Group Size: All treatments retain 15% of area Large groups — 0.8 – 1.2 ha Medium groups — 0.2 – 0.5 ha Small groups — up to 0.1 ha

Riparian Retention: All catchments retain 15% Group Removal: All treatments retain 15% Groups are retained along streams to represent Short cutting cycle — every 5 – 7 years 0%, 15%, 50%, and 100% of the stream length Long cutting cycle — every 20 – 30 years

Western Forest Strategy – AM Summary 59

Variable Retention Adaptive Management (VRAM) Experimental Sites Western Forest Products AM Program

Objectives

Variable Retention Adaptive Management (VRAM) experiments were designed by a team of scientists and foresters to address many of the questions regarding the layout and objectives for VR cutblocks. The objectives of these experiments are to:

1. Establish a series of designed comparisons of variable retention options to support an adaptive management approach; 2. Compare a range of retention levels and spatial patterns of retention in several forest types; 3. Monitor the short-term and long-term impacts of variable retention options on forest growth, structural attributes and selected forest-dwelling plant and animal species.

VRAM sites focus on the stand-level questions associated with biodiversity indicators 2 and 3 above (forest structure and species) as well as the silvicultural implications of VR options. Key questions include: What is the effect of the amount (percent retention level) and pattern of retention (dispersed trees, small groups, large groups)? What is the effect of size of opening and timing of adjacent openings? What is the effect of group location on small stream-riparian impacts?

Experimental Design

The experimental design for the VRAM areas consists of three replicates for each of five comparisons:

1. Group retention levels: 10%, 20% and 30% Group Retention; groups range in size from 0.2 to 0.5 ha. 2. Dispersed retention levels: 5%, 10%, and 30% Dispersed Retention; single trees to small groups up to 0.1 ha. 3. Group size: large groups (0.8 – 1.2 ha), small groups (0.2 – 0.5 ha), and dispersed trees (single trees to very small groups up to 0.1 ha); the retention level is 15% for all treatments. 4. Group removal – short/long cycle: group removal – short cutting cycle (5 – 7 years), group removal – long cutting cycle (20 – 30 years); groups in both treatments range in size from 0.1 to 1.0 ha. 5. Riparian retention: 0%, 15% and 50% of the length of small streams within treatments are covered by group retention (i.e., 0.25 ha or larger groups); a retention level of 15% is maintained within all stream catchments.

Each VRAM block consists of five 20 ha treatments: three VR options, clearcut and uncut. Each VRAM site requires an area of 80 to 100 ha that is as uniform as possible in timber type, plant associations and topographic features. Several cutblocks in close proximity can be used in place of a single contiguous area. Potential areas are evaluated for suitability by the science team. Forestry planners then complete a preliminary layout of treatment blocks. Treatments are randomly allocated to blocks before field layout commences. Regular communication occurs between the scientists and planners during layout to ensure that the objectives are met and the study design is not compromised. The original plan called for the fifteen VRAM blocks to be installed by six BC coastal timberlands units over a 7-year period as part of their annual harvest operations, with at least of one of the three sites for each comparison in an old growth forest type.

Each of the VR options must meet the requirements of the “retention silvicultural system”, which is the most common system used for implementing the variable retention approach. The retention system maintains enough tree canopy to have forest or residual tree influences on the majority of a cut area, and leaves long-term live and dead tree reserves of varying sizes and canopy layers, distributed throughout

Western Forest Strategy – AM Summary 60

harvested areas (Forest Practices Code of BC Act 1999). Mitchell and Beese (2002) describe the retention system and its rationale as a silvicultural system. To simplify the operational implementation of the retention system, our company guidelines state that retained groups of trees should be greater than 0.25 ha in size and no more than four tree lengths apart; individual trees or smaller groups should be no more than two tree lengths apart. Consequently, no place within a cutblock is more than two tree lengths from some standing trees.

The Group Removal comparison includes some additional specifications (see below). Each group removal treatment retains 15 percent of the area in long-term retention. Harvesting of groups takes place over three passes at the prescribed intervals. In conventional terms, the short- and long-cycle treatments could be considered group shelterwood and group selection systems. Each treatment must have at least 3 examples of the following size classes per pass: 1.0, 0.5, 0.25 and 0.1 ha (a total of nine each). Different sized groups must be intermixed throughout the block, not concentrated in one portion of the block. Other considerations for this treatment include: planning for permanent access skid roads, danger tree felling of snags in adjacent groups, orientation of groups to maximize light (north-south) and layout of groups to facilitate falling and yarding.

Western Forest Strategy – AM Summary 61

Variable Retention Monitoring and Adaptive Management Group Removal comparison blocks - specifications

Key question:

What is the effect of size of opening and timing of adjacent openings on regeneration, growth and yield, forest organisms, habitat attributes, forest health, windthrow and costs?

Comparison:

Group removal – short/long cycle: group removal – short cutting cycle (5 – 7 years), group removal – long cutting cycle (20 – 30 years); groups in both treatments range in size from 0.1 to 1.0 ha (see below).

Layout:

• The area should be as uniform as possible in timber type, site series and topographic features. Several cutblocks in close proximity can be used in place of a single contiguous area. Each block will have 4 treatments: clearcut, uncut (old growth or 2nd growth), and the two group removal alternatives (20 ha minimum size for each treatment, not including permanent roads). Each comparison set will require an area of at least 80 ha, with random allocation of treatments. • Each group removal treatment will retain 15% of the area in long-term retention following the spatial guidelines for VR (i.e., 4 tree length between retained groups of 0.25 ha or larger). • Each treatment will have at least 3 examples of the following size classes per pass: 1.0, 0.5, 0.25, 0.1 ha (9 each total). Different sized groups should be intermixed throughout the block, not concentrated in one portion of the block. • Harvesting groups will take place over 3 passes at the prescribed intervals.

Considerations:

• Plan for permanent access skid roads. • Danger tree felling of snags in adjacent groups. • Orientation of groups to maximize light (North-South) if possible. • Small openings may work best if they are longer in one direction to facilitate falling (e.g., 20 x 50 m). They may have to be somewhat larger to accommodate falling. • Size of openings should include any adjacent right-of-way opening.

# Groups/pass Size +/- 10% Ha/pass Total Ha 3 1 ha 0.1 3.0 9.0 3 0.5 ha 0.05 1.5 4.5 3 0.25 ha 0.025 0.75 2.25 3 0.1 ha 0.01 0.3 0.9 12 Total 5.55 16.65 Retention (15%) 3.0

W.J. Beese October 2000 VR-GroupRemoval.doc

Western Forest Strategy – AM Summary 62

VRAM Site characteristics descriptions (DRAFT)

Nine of the fifteen planned VRAM experimental sites were completed by 2005. Due to company ownership changes and economic conditions, no further VRAM sites have been established. General site location, forest type, elevation, harvest system and timing of harvesting are given in Table 1. More detailed descriptions of site characteristics follow [under construction]. Significant deviations from the study design or conceptual layout are also described.

Table 1. Characteristics and schedule for VRAM experiments

Elevation Harvest Comparison Location a Forest Type (m) System c Year b

Group MI – Tsitika R.* 1 – HwBaCw 550 Hoe 2001 retention % SW – Goat Isl. 1 – HwBaYc 600 Hoe/Cable 2004 QC – Hoodoo 2 – SsHwCw 100 Hoe 2004

Dispersed SW – Horseshoe Lk. 2 – Fd 100 Hoe 2002 retention % TBA TBA

Group size PM – Cluxewe R. 2 – HwBa 100 Hoe 2002 PA – Klanawa R. 1 – CwHwBa 300 Cable/Hoe/ 2005 TBA Heli

Group MI – Memekay R. 1 – HwBaCw 700 Hoe 2004 removal TBA TBA

Riparian SW – Lewis L. 2 – FdHw 500 Hoe/Cable 2003 retention MI – Moakwa Cr. 1 – FdHwCw 450 Hoe/Heli 2004 TBA

a MI = Mid Island Forest Operation (formerly North Island Timberlands), SW = Stillwater FO, QC = Queen Charlotte FO, PM = Port McNeill FO, PA = Port Alberni FO (formerly West Island Timberlands). *The Tsitika VRAM site is now within BC Timber Sales tenure. b 1 = Old growth, 2 = Second growth; Hw = western hemlock, Ba = amabilis fir, Cw = western redcedar, Yc = yellow cedar, Fd = Douglas-fir, Ss = Sitka spruce c Hoe = Excavator forwarding, Cable = grapple yarding, Heli = helicopter yarding

Table 2. Latitude and longitude for VRAM sites

Tsitika (50°20.3’N, 126°25.2’W) Goat Island (50°1.6’N, 124°29.3’W) Hoodoo (53°34.5’N, 132°6.5’W) Horseshoe Lake (49°54.8’N, 124°18.4’W) Cluxewe (50°33.1’N, 127°8.4’W) Klanawa Memekay Lewis Lake (49°57.8’N, 124°20.2’W) Moakwa (50°6.2’N, 126°5.2’W)

Western Forest Strategy – AM Summary 63

Riparian

Lewis Lake (UL806)

Treatments: Four streams were originally chosen for the study (L1, L2, L3, L4). Another stream (L5) was added after the first summer of monitoring. Streams L1, L2, and L3 were randomly assigned treatments, while streams L4 and L5 were designated as uncut treatments. Actual retention along the streams was: L1-0% (0%, 16.7% in catchment), L2-15% (16%, 17.6% in catchment), L3-50% (41%, 15% in catchment). Retention patches range in size from 0.6 to 2.8 hectares. Hand-falling began in April 2004; yarding by hoe-forwarding and grapple (cable) yarding completed by April 2005. Planted: Douglas-fir. Some windthrow in November 2007 storms; some salvage outside of treatments.

Treatments - estimated catchment areas (ha): 0% = 29.2, 15% = 25.5, 50% = 29.2, Uncut (100%) areas: __ha and ___ha [outside of UL806]; total UL806 area is 114.4 ha BEC: CWHdm Site series: predominantly zonal (01 Hw – Flat moss); Understory cover is near streams is dominated by salmonberry, Vacccinium spp., swordfern and devil’s club. Slope: moderate to steep Aspect: westerly (NW, W, SW) Elevation: 420 – 650 m Forest cover: second-growth Douglas-fir and western hemlock, approximately 80 years old, fire and logging origin

White River – Moakwa (OP 2071) Treatments: four streams were chosen for study (W1, W2, W3, and W4). The streams were randomly assigned treatments. One additional uncut (control) stream was established nearby on White River Main (now within Opening 10756-WR). Nominal retention along the streams was: W 1 (0%), W 2 (100%), W 3 (50%) and W 4 (15%); actual retention length and catchments to be determined. Hand-falling began in late June 2004; yarding by hoe-forwarding was completed by mid November 2004 (mostly by September). A small amount of helicopter yarding was completed in November 2005. Treatment areas (ha): to be determined. General location: TFL39, Block 2; Moakwa Creek (MC9) in the White River watershed, ~30km south of Sayward BEC: CWHvm1 Site series: predominantly zonal (01 HwBa – Blueberry) Slope: mostly gentle to moderate; minor steep Aspect: easterly (NE, E, SE) Elevation: 420 to 520 m; average 475 Forest cover: old-growth; Douglas-fir, western hemlock, western redcedar.

Group Retention %

Goat Island (GI-100)

Treatments: clearcut, 10% (actual 10.5%) and 20% (actual 21%) treatments occur together in the eastern portion of the study area without buffers between treatments; the 30% group retention block is to the west, surrounded by uncut forest at the time of harvest; groups range in size from 0.2 to 0.65 ha. Hand- falling began in late March 2004; yarding by hoe-forwarding and grapple (cable) yarding was completed by December 2004. Planted: Douglas-fir Significant windthrow in December 2006 storms.

Treatment areas (ha): Uncut 24.8, Clearcut 24.1, 10% Retention 19.9, 20% Retention 19.3, 30% Retention *.*.

Western Forest Strategy – AM Summary 64

General location: TFL39, Block 1; Goat Island (Powell Lake), ~20km north of Powell River, Spire Main (SP380, SP450) BEC: CWHdm (650m is the average elevation boundary of the CWHvm2, so a small portion of the clearcut could be in the vm2) Site series: predominantly zonal (01 Hw – Flat moss) Slope: * to * %, variable Aspect: varies Elevation: 500 – 670 m Forest cover: Second growth Douglas-fir, ___years old, following wildfire (?).

Tsitika (Op 38250)

Treatments: clearcut, 10% (actual 14%), 20% (actual 25%), 30% (actual 30%) group retention; groups range in size from ___ to ___ ha. Hand-falling began in mid July 2001; yarding by hoe-forwarding completed by April 2002. Planted: Douglas-fir Significant windthrow in November 2001 storms. Standing retention calculated in 2004 was: 10%=2%, 20%=7%, 30%=9%. No windthrow salvage has occurred.

Treatment areas (ha): Uncut __, Clearcut __, 10% Retention ___, 20% Retention ___, 30% Retention ___.

General location: TFL39, Block 2; Tsitika River (now BC Timber Sales area), ~__km north of Sayward; BEC: CWHvm1 Site series: predominantly zonal (01 HwBa – Blueberry) Slope: * to * % Aspect: Elevation: – m (550m average) Forest cover: old growth, western hemlock, amabilis fir, western redcedar

Hoodoo (HD216)

Treatments: clearcut, 10% (actual __%), 20% (actual __%) , 30% (actual __%) group retention; groups range in size from ___ to ___ ha. Hand-falling began in September 2004; yarding by hoe-forwarding completed by spring 2005. Planted: ___ Some windthrow in ______storms.

Treatment areas (ha): Uncut __, Clearcut __, 10% Retention ___, 20% Retention ___, 30% Retention ___.

General location: TFL39, Block 6; northeast Graham Island, QCI/Haida Gwaii, ~__km west of Port Clements BEC: CWHwh1 Site series: predominantly zonal (01 HwSs – Lanky moss) Slope: generally flat, * to * % Aspect: Elevation: – m (550m average) Forest cover: second growth; Sitka spruce, western hemlock, western redcedar and lodgepole pine; fire origin, high density

Group Size

Klanawa (Op 862222)

Treatments: clearcut, large groups (0.8 – 1.2 ha), small groups (0.2 – 0.5 ha), and dispersed trees (single trees to very small groups up to 0.1 ha); the planned retention level was 15% for all treatments (actuals were: large group – 14.7%, medium group – 14.5%, small group – 15.2%). Hand-falling began in mid

Western Forest Strategy – AM Summary 65

November 2004; yarding by grapple yarding and helicopter completed by February 2005. Planted: ___ Significant windthrow in ______storms.

Treatment areas (ha): Uncut 19.7, Clearcut 9.9, Small Groups 19.0, Medium Groups 16.6, Large Groups 14.0.

General location: TFL44; Klanawa River, north fork, NF170, ~__km south of Port Alberni BEC: CWHvm1 Site series: predominantly zonal (01 HwBa – Blueberry) Slope: moderate to steep, * to * % Aspect: Elevation: – m (300m average) Forest cover: old growth; western redcedar, western hemlock, amabilis fir

Cluxewe (Op 5599)

Treatments: clearcut, large groups (0.8 – 1.2 ha), small groups (0.2 – 0.5 ha), and dispersed trees (single trees to very small groups up to 0.1 ha); the retention level is 15% for all treatments. Hand-falling and hoe-forwarding of the clearcut and dispersed treatments was completed over the winter of 2002/03; the small / large group treatments commenced harvesting in Oct 2004 and were completed in early 2005. Planted: ___

Treatment areas (ha): Uncut ___, Clearcut __, Small Groups ___, Medium Groups ___, Large Groups ___.

General location: TFL39, Block 4; ~__km west of Port McNeill BEC: CWHvm1 Site series: predominantly zonal (01 HwBa – Blueberry) Slope: generally flat, * to * % Aspect: Elevation: – m (100m average) Forest cover: second growth; western hemlock and __; windthrow origin, high density

Dispersed

Horseshoe Lake (TM188)

Treatments: clearcut, 5% dispersed retention of single trees, 10% dispersed retention (similar spacing to 5% but with two trees at each location), 30% dispersed retention (small groups of trees up to 0.1 ha). Hand-felled in winter of 2001/2002, yarded by hoe-forwarding, and replanted in the spring with 85% Douglas-fir and 15% western redcedar.

Treatment areas (ha): Uncut 24.8, Clearcut 24.9, 5% Retention 26.3, 10% Retention 23.0, 30% Retention 24.6.

General location: TFL39, Block 1; south shore of Horseshoe Lake, ~15km northeast of Powell River BEC: CWHdm Site series: 90% zonal (01 Hw – Flat moss) Slope: mostly < 10% Aspect: varies; wraps around the northern half of a small hill Elevation: 170 – 260 m Forest cover: Second growth, ___years old, following wildfire. Study site species composition: 40% Douglas-fir (Pseudotsuga menziesii); 30% western hemlock (Tsuga heterophylla); 12% western redcedar (Thuja plicata); 15% red alder (Alnus rubrus), 3% pine (Pinus contorta).

Western Forest Strategy – AM Summary 66

Group Removal

Memekay 610

Treatments: Retention after the first pass at Memekay is 69% overall, calculated as follows:

First pass logging: 12.6 ha 2nd pass: 10.6 3rd pass: 10.6 Retention: 6.8 Total area: 40.6 (short- and long-cycle areas combined) Current retention: 28 / 40.6 = 69%

Treatment areas (ha): Gross VRAM area is 86.8 ha; Clearcut: 17.1 ha, Uncut OG control: 20.3 ha, Short- cyle: __ha, Long-cycle___ha. Wildlife tree patch (cave reserve): 8.8ha

General location: TFL39, Block 2; northwest of Campbell River BEC: CWHvm1 Site series: predominantly zonal (01 HwBa – Blueberry) Slope: mostly < 15% (check) Aspect: Elevation: * – * m (700m average) Forest cover: Old growth, western hemlock, amabilis fir, western redcedar, yellow-cedar

Western Forest Strategy – AM Summary 67

Appendix 3

Analysis of Ecosystem Representation

Notes on Information Needs D. Huggard, B. Beese February 2010

Goal The Western Forest Strategy (WFS) “coarse filter” goal or broad indicator for maintaining biological diversity is to represent the full range of ecosystems within the non-harvestable landbase to maintain lesser known species and ecological functions.

Objective This project evaluates the current status of this indicator by determining the proportion of each ecosystem that will likely remain unharvested on WFP tenure. A secondary objective is to assess whether the size, shape, age and spatial distributions of unmanaged areas are appropriate for sustaining biodiversity. We will characterize non-harvestable areas by ecosystem, using the hierarchy of ecosections, biogeoclimatic units (zones, subzones and variants) and site series to indicate the contribution of these areas towards ecological representation. Areas of deficiency will be described to help inform forest management planning and strategies for addressing these gaps in the future.

Study Area The WFS Adaptive Management program applies to Western Forest Products’ TFLs (6, 19, 25, 37, 39 and 44) and Forest Licences on Vancouver Island and the mainland coast. For purposes of this evaluation and plan, areas on the QCI/Haida Gwaii and Central Coast that are managed under Ecosystem Based Management (EBM) guidelines are excluded. WFP private lands are also excluded from the analysis. • Stillwater is only mainland area included; • Forest Licence FLA19231 (Nootka Sound) will be included, but other small FLs will not (lack of TEM); • Jordan River (TFL25) will also be excluded Some tenures continue to be in a state of flux due to government take-backs. This analysis will be a snapshot in time and will require future updates as boundaries change.

Zoning We will summarize ecosystem representation using the new WFP zoning scheme; a new Forest Strategy Zoning layer was recently completed by Sue MacDonald. It is based on (simplified) ecosections, modified VILUP zones (Enhanced, General, Special; including equivalents established for the Stillwater area) and BEC units (xm, dm and mm1 are grouped as “dry” units). We will not summarize by TFLs or company Forest Operations separately.

Ecosystems We’ll summarize by BEC variants within ecosections and by TEM site series. Because there are not too many ecosections, it should be fine just to intersect the ecosections and TEM site series layers, rather than doing separate summaries for TEM site series and for ecosection*variant. The first task will be to join the WFP TEM layers (currently TEM-C, TEM-E, TEM-W – for the former Cascadia, Englewood and Western areas) into a single coverage.

Forest Cover Forest cover will be used to stratify the landbase by productivity, age and the amount of edge influence. - Non-forest: excluded from all analyses; derived from the forest cover layer (and/or TEM). Includes water, alpine, wetlands

Western Forest Strategy – AM Summary 68

- Productive / non-productive (aka scrub): we will summarize results for both “all” forest area and for “commercial” forest. We will need a layer with “scrub” mapped (previously, this was mature forest with <210m3/ha for legacy Cascadia areas). Different management units may have slightly different definitions of “productive” forest. We will use whatever criteria were used in the latest TFL/TSA timber supply analysis and document any differences. Previous representation analyses also reported separately for high SI (>=25m at 50 years) forest, but we decided to skip this, since site series will give us rough productivity classes. - Edge: The percent of the non-harvestable landbase >50m and >200m from the nearest harvestable stand is the most useful measure of the spatial distribution of NHLB. This requires buffering out 50m and 200m from the combined harvestable + partially constrained landbase, to create a layer with 4 classes: harvestable, 0-50m, 50-200m and >200m into the NHLB. Notes: 1) This is edge adjacent to any potentially harvestable stand, whether or not it has already been harvested. 2) The edge buffers should not be created by buffering in from the NHLB, because this will also create edge around natural non-forest. - Age: will be summarized in the harvestable (aka THLB) and non-harvestable (NHLB) landbases to look for any gross differences. Age classes: <40, 41-80, 81-250, >250 years. Conclusion was not to further separate the 81-250 year class, because there is mostly little area in intermediate ages.

Non-harvestable Areas - Overall approach: it looks like the simplest way to accomplish the analysis is to use what's already been done to define THLB/NHLB for the latest TFL management plans, and update these as needed (which has already been done for several -- such as adding OGMAs, WHAs etc.). The 'build from scratch' option is problematic because of inconsistencies across legacy company datasets. We may have to live with some slightly different assumptions about operability and constraints , but our results will be more consistent with TFL plans.

- Protected areas: Generally excluded; however, any small parks that are completely enclosed by the tenure and were recently created (say after 1997?) can be included as part of the tenure, since they were basically created in response to the company’s (or predecessors’) management. ÆNeed to check that this is reflected in the GIS layers. If some info is missing from little embedded new parks, like TEM, we can just exclude them. Older parks and parks adjacent to the tenure are not included. It would be difficult to draw meaningful conclusions if we only had some adjacent parks included. We will summarize the percent of the relevant variants that are in Provincial Parks across the province in any case for comparison to the WFP landbase.

- Full constraints. These are listed more-or-less in the order that we want to overlay them. This order allows us to see what happens when we omit constraints at the bottom – usually softer, less certain constraints. For example, having draft OGMA’s low on the list lets us ask how much representation would change if these were taken away, without taking away harder constraints higher on the list like class 5 terrain where these overlap. - Class 5 terrain. Unstable terrain; minor operational adjustments occur but we will assume these areas are unharvestable. - Riparian reserves. Need to check what is available for mapping. Ideally actually reserves would be mapped, but we could also set up buffers of appropriate widths around different stream classes (assuming these are mapped). Note: Just the fully constrained reserves are included, not the wider management zones. - Ungulate ranges. Deer, elk, goats. - Wildlife reserves (WHAs) with full constraints (MAMU, NOGO, grizzly, others?). Need to include all fully constrained wildlife reserves. We should get the latest versions of these and include, even if some are still in a state of flux. - Established OGMA’s. Ones that are legally designated, or are pretty much sure to be. - Physically inoperable. Ideally this would not include any areas that are just economically inoperable, but that is sometimes hard to differentiate from physically inoperable. Operability lines will vary by tenure. We will use whatever criteria that were used in the latest management plan for the unit and document any differences among units.

Western Forest Strategy – AM Summary 69

- Mapped WTPs (WTRAs). These have not been mapped consistently across the tenures from a common start date. For this analysis, we’ll ignore them and include a blanket statement in the report that stand-level retention is not included (except any riparian reserves that are counted as stand-level retention). - Draft OGMA’s. These are at the bottom of the list, so that we can see what incremental effect they have, beyond areas that are already constrained.

- Partial constraints. We decided to ignore these in the analysis; however, there may be some that were included in the latest available dataset for a management unit. We will document any differences among units. We will mention that these areas provide an opportunity where representation can be increased with the least incremental cost.

Responsibilities Bill will co-ordinate the project and set up contracts with Annette under the FIA LBIP. Pat Bryant will assemble the GIS information after review of this document and discussion / clarification with Bill, and send data to Aaron. Dave will work with Aaron to finalize the steps in the GIS analysis and GIS outputs. Dave will prepare summary tables, graphs and a report on the results. Bill will include this report as a section in the overall Adaptive Management summary and plan.

Western Forest Strategy – AM Summary 70

Appendix 4

Western Forest Strategy Retention System Implementation Standards

Forest Strategy Working Group 6 February 2008; revised 12 June 2008

The following questions address key issues for implementation of the strategy.

1. How is the Retention System defined?

The term retention system refers to a specific silvicultural system designed to meet the goals of the variable retention approach. It was originally defined in the BC Operational Planning Regulations (March 1999) and has 3 requirements: 1) retention of trees distributed across the cutblock; 2) trees are left for the long term (at least one rotation); 3) distribution of leave trees achieves >50% “forest influence”. The specific definition of the retention system is:

“a silvicultural system that is designed to:

a. retain individual trees or groups of trees to maintain structural diversity over the area of the cutblock for at least one rotation, and c. leave more than half the total area of the cutblock within one tree height from the base of a tree or group of trees, whether or not the tree or group of trees is inside the cutblock.” The distribution of long-term retention over the area of the cutblock is open to interpretation, but the spatial requirement in “b” for “forest influence” provides the minimum standard for distribution. The retention system is no longer officially defined in BC legislation; however, the BC Forest Planning and Practices Regulation (Div.5, 64(4)) exempts harvesting from maximum cutblock size restrictions when certain spatial requirements are met for group or dispersed retention. The retention system is considered a “partial cutting” approach and is categorized as an “even-aged” system despite the resulting uneven- aged forest because the cut areas are regenerated and managed much like other even-aged systems. For further description see Mitchell and Beese (2002) or the on-line Introduction to Silvicultural systems at: http://www.for.gov.bc.ca/hfp/training/00014/chap2frt.htm

2. What are WFP’s goals for cutblocks using the Retention System?

A. To design and implement the retention system in a safe and cost-effective manner. B. To leave a biological legacy of attributes from mature and old forests, well distributed within stands and landscapes, to maintain and promote biological diversity within the company’s public tenures. C. To design cutblocks to maintain forest influence on the majority of the harvested area throughout the rotation. D. To ensure that cutblocks meet the principles of forest stewardship (i.e., prescriptions address silviculture, forest health, site productivity, visual aesthetics and other values).

Western Forest Strategy – AM Summary 71

3. How much long-term retention is required in Retention System cutblocks?

The minimum retention required in each Retention System cutblock by VILUP Resource Management Zone is: • Enhanced: 10% • General: 15% • Special: 20%

This retention is long-term—it must remain for at least one rotation. Salvage for windthrow, fire, insects and disease is allowed only for amounts left above these minimums for the block to qualify towards WFP’s retention system targets. In some cases, it may be more desirable to salvage all damaged trees and re-classify the block as a clearcut; however, it will no longer count towards targets. [Note: there is no longer a 5% dispersed retention minimum for Enhanced zones as there was under former “Timber” zones.] The zone targets do not apply to clearcut with reserves (CCR). Stand-level retention for CCR cutblocks is guided by WTRA (Wildlife Tree Retention Areas, formerly WTP) requirements (3.5% minimum per cutblock and 7% average for all cutblocks harvested in a year, unless other percentages are specified in an FSP or SFM plan). There is a 5% increase in the Enhanced and General Zone targets for drier BEC subzones/variants (CWHxm, dm, mm1).

4. How is % retention calculated?

The % retention targets represent a ratio of the area retained to the area logged: ______ha retained______ha logged (NAR + NP roads) If the target is 10%, then for every 40 ha logged (stumps, landings and roads) we leave 4 ha of retention. A group retention calculation should be on file (see example below). Data will be extracted from GENUS (aka CENFOR) when fully implemented.

Example of Group Retention Determination

WFP GROUP RETENTION CALCULATION

Zone: Enhanced: Minimum Retention Requirement Minimum (39 ha cut area, NAR+NP roads) 3.9 ha Retention %: 10%

1. Total Area (ha) of Group Retention: 4.2 ha (Including: WTRAs, RRZs, and other group retention)

2. Total Area Equivalent of Additional Long-term Dispersed Retention 0.4 ha (basal area of retained trees / total pre-harvest b.a. per ha)

TOTAL RETENTION (ha) 4.6 ha (%=4.6/39 x 100) (12%)

Note that this calculation results in a lower amount of retention required for a given percentage compared to using the total area under prescription (TAUP).

Western Forest Strategy – AM Summary 72

5. How is forest influence calculated?

• Forest influence is calculated as: All long-term retention area + the cut area within 1 tree length of standing timber Total Area Under Prescription (TAUP)

• Cut area is NAR + NP roads; long-term = retained for at least one rotation. • The tree length used should be the average co-dominant/dominant height for the retained trees or cutblock boundary segment providing influence. For example, you may cut 40 m tall trees in the block, but if the cutblock boundary has 20 m tall trees, you must use 20 m for the influence calculation. Adjacent areas must be “free growing” to count the edge segment for influence. There is a forest influence calculation tool in ArcView Planner which enables buffering one tree height around retention and cutblock edges. [This will be updated to meet the new method of influence calculation.] • Some calculation of forest influence should be on file to demonstrate compliance with the retention system definition. Because blocks may have sufficient retention that >50% forest influence is not in doubt, a detailed calculation is not always necessary. In these cases a rough estimate will suffice. • Forest influence is a “rule”, because it defines the retention system; therefore, there is no flexibility around the >50% figure. While single trees and cutblock edges do not provide the same degree of actual influence, for simplicity, both contribute equally to meeting the influence rule.

6. Does WFP have spatial guidelines for ensuring that the structural diversity over the area of the cutblock requirement is met?

Yes. To ensure consistent application of the Retention System across the company, WFP will apply the following spatial guidelines:

• At least 5% of the minimum retention requirement should be clearly visible within the cutblock (i.e., “internal” retention). For example, if you are planning a cutblock in the General zone with a 15% retention requirement, 5% should be “internal” and the remaining 10% can be on the edge of the cutblock. Internal retention includes: patches, groups, single trees or clumps of trees that are non-contiguous with cutblock boundaries*; riparian strips crossing the block; and peninsulas of trees extending 1.5 tree lengths or more into the block. (Typically, a peninsula will extend further into the block than it’s width at the base; not just a big bend on the block edge.) *Exception: if a group is on the block boundary but will be surrounded by younger forest after logging (i.e., it will be visibly ‘internal’ to the harvested area), then it can also be counted towards the internal target. • Retained groups should be at least 0.25 ha whenever possible. In high windthrow hazard areas, group size should be increased. • Openings within cutblocks should be less than four tree heights across as a general guideline, not a strict “rule”. Layout should fit with the biological objectives, regeneration objectives, harvesting equipment constraints, and other engineering considerations. Windthrow hazard is one example of where the size of open areas within blocks may be adjusted to reduce potential for damage. • For safety and access reasons, retention should be avoided within 1.5 tree lengths of roads, particularly single trees and retention above roads. • Where individual trees or small clumps of trees dominate a block the system is called Dispersed Retention. As a guideline retained individual trees should be no more than two tree-lengths apart to provide the necessary influence and ecological structure throughout the block. When

Western Forest Strategy – AM Summary 73

individual trees are left uniformly throughout the block, normally they will be less than two tree- lengths apart to achieve the retention target for the zone (unless they are all very large trees). Where small clumps of trees are left, occasionally more than two tree-lengths may be necessary between clumps to successfully achieve all management objectives in a safe manner. In most cases, dispersed trees will be used as a supplement to Group Retention, or combined as Mixed Retention. In these situations, the two tree-length guideline does not apply. Avoid dispersed retention on sites where shade-intolerant species are preferred (e.g., Douglas-fir), and on slopes that are greater than 30%. • Locate retention on areas of high ecological value or “anchors”, such as riparian reserves, rock outcrops, wetlands, ravines, clusters of snags, or other areas of concentrated high structural diversity. Where a species has been identified as "at risk" other specific attributes may be desired and will help locate potential groups (e.g., large-limbed trees for Marbled Murrelet nesting sites). In addition to incorporating riparian reserves into stand-level retention, locate group retention wherever possible to provide protection for small headwater streams and wetlands that do not require buffer strips.

7. How are % retention system targets by Ecosection and Zone calculated and tracked?

• The retention system target (%RS) will be applied on an area basis (hectares logged) rather than a volume basis. Shelterwood and selection systems with long-term reserves also count towards the targets. • Each operation should plan to meet the targets by Ecosection and VILUP Zone within their area to the extent possible; however, tracking will occur on a corporate-wide basis. Some operations may have only a small portion of an Ecosection/Zone combination and may only have a single cutblock in a given year; therefore, applying annual targets to local operations is inappropriate. • To maintain flexibility, given the need to adjust harvesting schedules for weather, markets and other conditions, the %RS goals will allow annual variance of 25% from the target. Example: for an Enhanced zone in the Northern Island Mountains Ecosection, the % of annual harvesting using the RS should stay above 37% (50% x 0.75). • Forest Operations will record the area harvested each year by silvicultural system using a spreadsheet distributed by Corporate Forestry (or using a CENFOR/GENUS report, where applicable). Achievement of targets will be assessed for each Ecosection/Zone combination. Areas requiring adjustment among operations in order to achieve targets will be identified. • Targets may be refined by biogeoclimatic unit for some landscape units after a spatial analysis of representation of ecosystem types in Old Growth Management Areas (OGMAs) and other landscape-level reserves. Stand-level retention may be used to compensate for deficiencies in landscape-level representation of specific ecosystems. Extensive logging history and poor representation in reserves in leeward, drier biogeoclimatic variants (CWHdm, xm, mm1) is the rationale for setting targets that are higher for both the retention system and average percent retention for cutblocks in these variants. The overall target for the Zone/Ecosection, however, remains the same. • Targets do not apply to private land or TLs outside of a TFL. • This is a forward-looking strategy. Targets for the amount of the retention system used by zone apply to current and future harvesting plans. Operations will not return to previously logged retention blocks and remove reserves, except as allowed under salvage guidelines developed by the FSWG.

8. How will Phase-In be measured?

Targets for the retention system will be phased-in from 2008-2010 under the following schedule:

Western Forest Strategy – AM Summary 74

YEAR TARGET (% OF HARVESTED AREA) 2008 At discretion of each Region/Forest Operation 2009 Half way towards targets 2010 Full implementation

Although each operation should attempt to meet the schedule, we will measure WFP’s phase-in progress by Region to allow for greater flexibility. The guiding principle with phase-in is a desire to avoid changes to cutblocks already planned, unless there are financial incentives to do so.

9. Can the same cutblock have two silvicultural systems?

Yes, but this should not happen often. Preferably, an entire cutblock should be classified as the same silvicultural system (i.e., it either meets the retention system, or it is a clearcut with reserves). However, it is acceptable to split a block if there is a logical boundary and reason for it, such as: • Two portions of a block separated by a riparian reserve, where different systems are justified; • A significant slope break (usually at a road), where one side of the road is cable yarding and the other side is hoe-forwarding; • Different susceptibility to windthrow (a block wrapping around a mountainside); • Important ecological differences or resource features that justify different treatments.

In other words, no “lines in the slash” separating different units just to make part of a block count as the retention system.

10. How should retention be identified on our maps and in CENFOR (GENUS)?

Retention associated with a cutblock should have one of the following identifiers:

1. WT = Wildlife Tree Retention Area (formerly WTP). Use this category for areas designated to meet legal WTRA requirements. 2. LT = Long-Term Retention. Use this category for all other long-term retention (i.e. unavailable for harvest within the rotation, and intended to be left in perpetuity). 3. ST = Short-Term Retention. Use this category for any other retention available for harvest within the rotation but with some time constraint on removal (e.g., areas left to meet other purposes such as silvicultural systems (shelterwood, selection), visuals, adjacency and timber type).

Use the following sub-categories for WT and LT:

r = riparian reserve zones only; other riparian areas such as riparian management zones are reported as "o" (or “u”, if applicable). o = operable/harvestable; normally would be available for harvesting but designated for WT or LT retention. u = unharvestable; not available for commercial harvesting now or in the future due to operability constraints, low site productivity, unstable terrain, other values. (There may be some opportunity for standing stem harvesting.)

Western Forest Strategy – AM Summary 75

We will use these designations for mapping and tracking retention in our various WFP systems.

Example: A cutblock that has riparian reserves that are partially designated as WTRA would label the reserve zone as WTr (for the WTRA) and LTr (for the rest of the reserve not designated as WTRA). Additional retention patches left to meet the retention system requirements would be labeled LTu or LTo. Patches of younger forest within the cutblock that can be harvested before the end of the next rotation would be designated as ST.

Terms such as “TLA=Timbered Leave Areas”, “VR”, “Reserves” will not be used for identifying stand-level retention. Details on reporting to government in RESULTS and tracking in CENFOR are under development.

It is important to identify all of the retention associated with the cutblock or TAUP that fits within a logical management unit. This includes areas such as stream buffers or gullies at the edge of a harvested area that will not be accessible to future harvesting. For inventory and cut control purposes, we need to identify such “default” retention so that it is not assumed to be available in the THLB and thus contributing to AAC. There may, however, be patches of timber between different units of a block that are available for future harvesting and have no constraint on removal; these areas are not retention, so they do not need an identifier.

11. When should a cutblock be “claimed” for tracking %RS targets? What if a block straddles a zone?

Enter a cutblock on the tracking spreadsheet as layout is completed. Record the date when falling begins (month/year is sufficient), and record the year when the cutblock is completed (primary logging). This will avoid partial claims for blocks spanning two years and potential double-counting. Once the strategy is fully implemented, we will track our progress using harvest completion. During phase-in, the tracking spreadsheet will also allow us to document our progress using completed layout and cutblocks underway.

For most blocks straddling a zone boundary, use the retention targets for the zone that comprises the majority of the harvested area and claim the block accordingly. If it is a large block that is equally represented in two zones (probably a rare occurrence), ensure that both zone targets are met and claim the hectares logged for each zone.

12. What training will be done?

Efficient and safe implementation of the retention system in operations that have had little experience with the approach will require training. The following training activities have occurred or are planned for 2008: • A PowerPoint slide show was developed to illustrate the wide range of approaches to the retention system, using examples of cutblocks throughout the company. (Dec 31, 2007) • Retention system implementation standards were developed by the FS Working Group and distributed to operations (February). This document was updated with standards for mapping and tracking, and distributed with a tracking spreadsheet (June). More detailed guidelines for the retention system will be developed by the FSWG (Third Quarter). • W.J. Beese and FSWG members conducted training sessions on retention system layout, guidelines and mapping at Holberg, Jeune Landing, Nootka Sound, Gold River and Jordan River operations (May 2008). Meetings were also held at Pt. Alberni, Pt. McNeill, Englewood to ensure consistent application of the retention system across the company (May-June; remaining operations will have meetings in July). • Tailgate Training Posters were used in training sessions, updated and distributed to each operation (PowerPoint-June, Posters-July). • The Regional Engineers and Foresters will organize job exchanges for a week or more between legacy company field planning staff with different levels of experience with layout for the retention system (Second Quarter 2008).

Western Forest Strategy – AM Summary 76

Appendix 5

RESEARCH PROJECT SUMMARY Impacts of variable retention on small streams and riparian areas

Note: This study began in 2001/02 with funding by Forest Renewal BC. It was supported in 2002/03 by Forestry Innovation Investment (successor program to FRBC), then dropped as a result of funding cutbacks the following year. Stream monitoring continued with company funding through 2008 on two VRAM sites (Lewis Lake, Moakwa). Several students were supported as part of the project. Interfor has continued monitoring the Kennedy site under the Forest Investment Account.

Rationale

Retention of standing trees along stream channels during forest harvest has been widely implemented as a method of minimizing impacts on fish populations. As a result, requirements or guidelines for forested buffers have largely been limited to larger streams with significant populations of fish. The effects of forest harvest on small streams are less well known. Even more poorly understood are the cumulative impacts alterations in the functions of multiple small streams have on larger, downstream reaches. Headwater streams may be important for the production, processing and downstream export of nutrients, organic matter, and stream invertebrates, thereby maintaining the productive capacity of downstream fish habitat.

Included within the “small-stream” category are the smallest fish-bearing streams (Forest Practices Code Class S4), and streams without fish (primarily Forest Practices Code Class S6). Collectively, these low- order streams can comprise 75% or more of the total length of the drainage network within coastal watersheds; therefore, there are concerns about harmful impacts from harvesting on riparian and aquatic resources within the drainage network downstream. VR is fundamentally different from the application of a narrow buffer along the stream channel, an approach most commonly used in forest harvest to protect streams. No evaluation of a discontinuous patch approach to streamside management, like VR, has been evaluated. However, this type of management option is being applied both on Weyerhaeuser lands in BC and along streams without fish on all privately owned forest lands in Washington, suggesting that application of this approach may become more widespread. In addition, much of the research evaluating streamside buffers or other management options has focused on larger streams.

Our knowledge of the effect of forested buffers is currently being extended to small stream channels with work ongoing at several locations in B.C and Washington. (DFOs Stuart/Takla project, BC; J. Richardson’s Malcolm Knapp Forest study, BC; R. Bilby’s Upper Green River study, WA). There are no studies currently underway which evaluate the relative effectiveness of a discontinuous patch approach like VR. Our project will provide this information and compliment the ongoing studies on buffers.

Key Questions Addressed

1. To what extent does VR moderate temperature changes caused by forest harvest? Do temperature changes affect the rates of organic matter processing, stream productivity, and the growth and life-cycles of stream invertebrates? Do changes in processing rates and seasonal timing influence fish communities occupying the study site or downstream stream reaches? 2. What are the short and long-term effects of VR on woody debris supply? What are the consequences for channel morphology, habitat complexity, and material transport? What is the effect on algae, invertebrates, or fish inhabiting the stream?

Western Forest Strategy – AM Summary 77

3. What are the effects of VR on nutrient input, cycling and transport in headwater streams? 4. To what extent does VR alter litter input and primary production? Do these changes shift stream productivity and food webs? What are the impacts on the quality, quantity, and timing of downstream nutrient and organic material transport?

Study Objectives

1. Determine the extent to which clear-cut harvest along small streams alters the physical, chemical and biological attributes of these systems. 2. Determine the extent to which changes associated with clear-cut harvest are moderated by two variable retention (VR) options for streams. 3. Determine the time for recovery to pre-harvest conditions for a variety of system attributes under clear-cut and VR harvest options. 4. Use the information generated in addressing the preceding objectives to generate predictions of the cumulative effects of timber harvest along small streams under various management scenarios on downstream reaches.

Experimental Design and Treatments

This is a multi-year experiment comparing the effectiveness of different variable retention patterns along small stream channels < 3 m wide. Specifically, these are class S4 (fish-bearing) and S6 (fishless) streams where riparian reserves are not mandatory and logging to the stream bank may occur. The streams are headwaters emerging as transitional or 1st order streams from the study watersheds. The experiment consists of four treatments:

1. Clearcutting up to both stream banks (no retention within 25 m of either stream bank); 2. Intermediate level of retention, with patches covering at least 20 m on both stream banks along approximately 15 – 20 % of the channel length.; 3. Streamside focus for retention, with patch retention concentrated along the stream, protecting approximately 50% of the channel length; 4. Unharvested control.

Two sites were chosen from the company’s VR Adaptive Management (VRAM) study areas (near Powell River and Sayward) and a third site was located in Interfor’s operations near Ucluelet. Each study area was delineated into 4 treatment units, each aligned as closely as possible around the stream catchment boundaries (about 10 to 30 ha in size). As much as is possible the streams were matched for gradient and size. Treatments were allocated randomly to the units. A 15 to 20% retention level was applied to all of the harvesting treatment blocks using the retention system definition for forest influence and spatial distribution. Group retention patches were 0.25 ha or larger, to ensure a buffer of at least 20 m of uncut forest on either side of the stream within the patch. Harvesting occurred in 2003 and 2004. Changes in measured attributes at harvest sites were compared before and after cutting and any observed responses were evaluated relative to the behaviour of that parameter at the uncut reference stream in that block.

This study is planned for 5 years, which will enable us to accurately characterize the short-term response to the various harvest options and begin to evaluate the rate of recovery to pre-harvest conditions for the measured attributes. Obviously, some of the measured attributes, such as wood input, would not be expected to return to pre-harvest conditions within the time frame of this study. However, there may be significant changes within the 2 to 3 years post-harvest period for some parameters (e.g., nutrient export, water temperature). Ideally, this study should continue for at least a decade, and the participants hope to secure additional support to this end after the initial phase of the project concludes.

Measurements

Physical, biological and chemical aspects of the streams draining the experimental sites were evaluated. The measurements fall within three categories: 1) Delivery of materials to the stream; 2) Characteristics of the stream at the site; 3) Export of materials from the downstream end of the treated site. Understanding

Western Forest Strategy – AM Summary 78

the changes in delivery of sediment, wood, organic matter and nutrients from the watershed to the stream provides a context for interpreting changes in characteristics of the channel that are observed. These downstream impacts may include changes to food availability and utilization in fish-bearing reaches of streams further downstream.

Weather and stream discharge affects many watershed processes, so weather stations were installed at each monitored VR block measuring precipitation and air temperature. As precipitation measurements must be taken in an open location, the weather stations were located at the clear-cut site or adjacent open areas. The expense of installing, rating and maintaining continuously recording discharge stations precludes locating such a device at each site. Discharge at each site was estimated by assuming that flow contribution is proportional to watershed area.

Cooperators

Bill Beese Weyerhaeuser, BC Coastal Group Bob Bilby Weyerhaeuser, Western Timberlands Max Bothwell Environment Canada, National Water Research Institute Dan Hogan BC Ministry of Forests, Research Branch Erl MacIsaac Dept. of Fisheries and Oceans Dan Moore University of British Columbia John Richardson University of British Columbia Peter Tschaplinski BC Ministry of Forests, Research Branch Warren Warttig International Forest Products (Interfor)

ABSTRACT From final report to Forestry Innovation Investment, April 2003 FII Reference No: R02-37 Variable retention and the conservation of small streams Authors: Bill Beese, Bob Bilby, Max Bothwell, John Heffner, Dan Hogan, Donovan Lynch, John Richardson, Jon Sleeman and Peter Tschaplinski

Small headwater streams receive little protection under forest practices guidelines, yet such streams may be important for maintaining the productive capacity of downstream fish habitat by processing and exporting nutrients, organic matter and stream invertebrates. The current study was initiated to evaluate the response of small streams to the new Variable Retention (VR) approach to forest management being applied throughout coastal BC forests. Three sites, one at Lewis Lake near Powell River, one in the White River drainage near Sayward and the other near Kennedy Lake were chosen for the main study. Several other streams in the Tsitika River drainage are being studied as part of a PhD thesis. Of the three sites, Kennedy Lake is unique in its geomorphological features, chemical composition and biological community structure. Pre-logging surveys of streams at all the sites identified disparities between adjacent, similar-looking streams both in nutrient chemistry and benthic communities, emphasizing the need for extensive pre-treatment data sets. High variability in nitrate levels both within and between streams at the Lewis Lake and White River sites preclude any generalization about potential for P or N limitation. In contrast Kennedy Lake streams appear to be more consistently N-limited. Extremely low flows during the late summer/early autumn led to very low concentrations of dissolved organic carbon (DOC; 1 mg/L–2 mg/L) at the Lewis Lake and White River sites and UV impacts may be observed following thinning of streamside canopies at those sites. On the other hand, DOC levels at Kennedy Lake were sufficiently high (10 mg/L–20 mg/L) to block UV. Low rainfall also led to intermittent flow in many streams with disconnected pools. Algal and bacterial biomass was low in all streams but algal communities were consistently higher and bacterial numbers lower in Lewis Lake streams compared to streams at White River.

Western Forest Strategy – AM Summary 79